Politics of Change:
Energy Efficiency Policy in Britain and Germany
Jan Rosenow
Thesis submitted in partial fulfilment of the requirements for the
degree of Doctor of Philosophy at the University of Oxford
St Hilda’s College
Oxford, Hilary Term 2013
Abstract
Abstract
Britain and Germany are two countries that are internationally recognised for their energy
efficiency policies in the area of domestic buildings. Although pursuing similar objectives,
the two countries have quite different flagship policies: Germany put in place a large loan
and grant scheme to finance residential energy efficiency measures in 2001, the CO2Building Rehabilitation Programme. The UK was the first country in Europe that
introduced Energy savings obligations in 1994. Since then the policy instruments in both
countries experienced significant change. This thesis explores the politics of the changes
that occurred, and investigates the policy processes that led to the modifications.
Mainstream theories of policy change emphasise short-term crisis-like events when it
comes to explaining why policies change significantly. However, more recent theoretical
approaches suggest that gradual mechanisms, including accumulating external pressures
and slowly developing consequences of the policies themselves, play an important role as
well. In order to approach the two cases theoretically, this thesis draws on the concept of
friction developed in Punctuated Equilibrium Theory and the policy feedback literature.
This thesis investigates how long-term pressures, for example the perceived impact of
climate change and rising energy prices, affected the evolution of the key home energy
efficiency policies in Britain and Germany. Combined with assessing the impact of
institutional changes and policy feedback, a comprehensive analysis of long-term policy
change is carried out. A set of different methods is employed to undertake the investigation
including qualitative and quantitative research methods such as semi-structured interviews
with more than 25 experts followed by qualitative content analysis, complementary
document analysis, and the review of data sets. The analysis provides a detailed historical
case study of the key home energy efficiency policy instruments in Germany and the UK
with a focus on the causal mechanisms of gradual pressures. Wider conclusions are drawn
for the theories of policy change and how gradual pressures might be accounted for in
those theories in a more meaningful way.
i
Preface
Preface
Acknowledgements
This research was supported by scholarship from the UK Energy Research Centre and the
Heinrich Boell Foundation. I am very grateful for this support and without it this research
would not have been possible.
My supervisors, Nick Eyre and Miranda Schreurs, not only helped me with the design and
the execution of the research, but provided critical feedback on my work and strengthened
the argument throughout the process. During times of low creative energy they often
injected fresh ideas and put things into perspective. Being supervised by two such
knowledgeable and inspiring people was an invaluable experience.
I would also like to thank the Lower Carbon Futures Group at the Environmental Change
Institute of the University of Oxford. It has been a pleasure to work in such a diverse group
of people all dedicated to energy demand issues. Having had the opportunity to see Gavin
Killip finish his doctoral thesis at a time when I started mine helped me a great deal in
understanding what the (admittedly sometimes painful) process of completing a doctorate
involves. Sarah Darby, Matthias Fripp, Yael Parag, and Brenda Boardman, my internal
examiners at different stages, deserve thanks for raising critical issues early on.
I owe a special thanks to the Electricity Markets & Policy Group at the Lawrence Berkeley
National Laboratory, where I spent the summer of 2012. I learned a lot about US energy
efficiency policy and mountain lions, although two and a half months without rain made
me question my decision to settle in Oxford.
Furthermore, I would like to thank Ray Galvin for writing a joint paper with me and the
Think Tank 30 of the German Chapter of the Club of Rome for offering me the
opportunity to disseminate my research results to a wider audience.
Without the many individuals who agreed to be interviewed and in this way contributed to
this research, this thesis would not have been completed. I am very grateful for their time
and comments.
With a project that takes as long to complete as a dissertation, I would not have been able
to finish this research without the support of my family and friends. Even though they
probably never quite understood why I was putting myself through this process, they
always showed their support. The birth of my son, Joshua, certainly forced me to work
efficiently, even though the lack of sleep did not help. The person who probably suffered
most from my DPhil adventures was Doerthe, who endured through my many theoretical
and methodological struggles. She always encouraged me to keep focused and put things
into perspective. I look forward to opening that bottle of Beerenauslese and sharing it with
her.
Jan Rosenow
Oxford, March 2013
ii
Preface
Publications
Parts of this thesis have been or are going to be published as academic papers:
Journal articles (peer reviewed):
Rosenow, J., Eyre, N. (in press): The Green Deal and the Energy Company Obligation. Invited paper to
appear in Proceedings of the ICE - Energy
Rosenow, J., Platt, R., Flanagan, B. (2013): Fuel poverty and energy savings obligations in the UK. Energy
Policy
Rosenow, J., Galvin, R. (2013): Evaluating the Evaluations: evidence from energy efficiency programmes in
Germany and the UK. Energy & Buildings 62, pp. 450-458
Rosenow, J., Eyre, N., Rohde, C., Buerger, V. (2013): Overcoming the upfront investment barrier comparison of the German CO2 Building Rehabilitation Programme and the British Green Deal. Energy &
Environment 24 (1&2), pp. 83-103
Rosenow, J. (2013): The Politics of the German CO2-Building Rehabilitation Programme. Energy Efficiency
6 (2), pp. 219-238
Rosenow, J. (2012): Energy Savings Obligations in the UK – A History of Change. Energy Policy 49, pp.
373–382
Journal articles (not peer reviewed):
Rosenow, J. (2013): A short history of Energy Saving Obligations in the UK. Energy Action 116
Rosenow, J. (2012): Will the Green Deal revolutionise UK housing? Energy World 410, pp. 33-34
Book chapters:
Rosenow, J. (2013): Energy efficiency policy in Germany and the UK. In: Urban, F., Nordensvard, J. (Ed)
Low Carbon Development: Key Issues. Earthscan, Oxford.
Conference papers:
Rosenow, J., Croft, D., Eyre, N. (2013): Energy policy in transition: evidence from energy supply and
demand in the UK. Proceedings of ECEEE Summer Study 2013. pp. 439-447
Rosenow, J. (2012): Fuel poverty and energy savings obligations: Natural bedfellows or enemies? Fuel
poverty in policy and practice - a postgraduate symposium. University of Sheffield, 16 November 2012.
Rosenow, J., Eyre, N. (2012): The Green Deal and the Energy Company Obligation - will it work?
Proceedings of British Institute of Energy Economics 9th Academic Conference. European Energy in a
Challenging World: The impact of emerging markets. St Johns College, Oxford, 19-20 September 2012.
Rosenow, J. (2012): Understanding policy change: the case of energy savings obligations in the UK.
Proceedings of World Sustainable Energy Days 2012. Wels, 28 February 2012.
Rosenow, J. (2011): An ecological economics approach to energy efficiency. UK Energy Research Council
Sparks Network, PhD Symposium. Imperial College, 20 October 2011.
Rosenow, J. (2011): Different Paths of Change: Home Energy Efficiency Policy in Britain and Germany.
Proceedings of ECEEE Summer Study 2011, pp. 261-272.
iii
Contents
Overview of contents
1
Introduction ..................................................................................................................... 1
2
Theories of policy change: friction and policy feedback .............................................. 10
3
Methodology and research design ................................................................................. 48
4
Case I: Supplier Obligation in the UK .......................................................................... 80
5
Case II: CO2 Buildings Rehabilitation Programme in Germany ................................. 169
6
Evaluation .................................................................................................................... 214
7
Cross-case analysis: politics of change ....................................................................... 238
8
Conclusions ................................................................................................................. 264
9
Bibliography ................................................................................................................ 275
iv
Contents
Detailed Table of Contents
1
2
Introduction .......................................................................................................................................... 1
1.1
Subject of investigation .......................................................................................................... 2
1.2
Theoretical approach: a brief introduction ........................................................................ 3
1.3
Importance of the domestic sector for climate change policy ..................................... 4
1.4
Research questions and aim ................................................................................................... 5
1.5
Approach ..................................................................................................................................... 6
1.6
Organisation of the thesis ....................................................................................................... 7
Theories of policy change: friction and policy feedback ..................................................... 10
Theories of policy change ................................................................................................... 11
2.1
2.1.1
Incrementalism .................................................................................................................. 12
2.1.2
Advocacy Coalition Framework (ACF) ..................................................................... 14
2.1.3
Multiple streams (MS)..................................................................................................... 16
2.1.4
Punctuated Equilibrium Theory (PET) ....................................................................... 19
2.2
General critique ...................................................................................................................... 22
2.3
A new lens on policy change.............................................................................................. 26
2.3.1
Friction ................................................................................................................................. 26
2.3.2
Policy feedback ................................................................................................................. 31
Combining punctuated equilibrium theory and policy feedback literature........... 36
2.4
2.4.1
Compatibility of friction and policy feedback literature ....................................... 36
2.4.2
Conceptual framework and hypotheses ...................................................................... 41
2.4.3
Hypotheses .......................................................................................................................... 43
Conclusions ........................................................................................................................................... 46
3
Methodology and research design .............................................................................................. 48
Ontological and epistemological assumptions .............................................................. 49
3.1
3.1.1
Ontology .............................................................................................................................. 49
3.1.2
Epistemology ..................................................................................................................... 50
Case study approach ............................................................................................................. 52
3.2
3.2.1
Potential pitfalls of the case study method ................................................................ 53
3.2.2
Within-case analysis as a means to avoid pitfalls.................................................... 55
3.3
Systematic process analysis ................................................................................................ 56
v
Contents
Selection of cases................................................................................................................... 57
3.4
3.4.1
Geographical scope .......................................................................................................... 58
3.4.2
Time horizon ...................................................................................................................... 59
3.4.3
Unit of analysis .................................................................................................................. 60
Research methods .................................................................................................................. 63
3.5
3.5.1
Critical assessment of qualitative research methods .............................................. 63
3.5.2
Document analysis............................................................................................................ 65
3.5.3
Interviews ............................................................................................................................ 66
3.5.4
Qualitative content analysis ........................................................................................... 72
Operationalisation.................................................................................................................. 74
3.6
Conclusions ........................................................................................................................................... 79
4
Case I: Supplier Obligation in the UK ...................................................................................... 80
Background and main changes of the Supplier Obligation ....................................... 80
4.1
4.1.1
Basic architecture of the SO .......................................................................................... 81
4.1.2
Main changes of SO from 1994 to 2010 .................................................................... 83
Analysis of policy drivers ................................................................................................. 101
4.2
4.2.1
Market liberalisation ...................................................................................................... 102
4.2.2
Change of key personnel .............................................................................................. 110
4.2.3
Transfer of target-setting powers ............................................................................... 116
4.2.4
Energy prices .................................................................................................................... 121
4.2.5
Climate change ................................................................................................................ 131
4.2.6
Fuel poverty ...................................................................................................................... 142
4.2.7
Energy efficiency industry ........................................................................................... 154
4.2.8
Perceived success of policy instrument .................................................................... 162
4.2.9
Policy drivers not found relevant ............................................................................... 167
Conclusions ......................................................................................................................................... 167
5
Case II: CO2 Buildings Rehabilitation Programme in Germany ..................................... 169
Background and main changes of CBRP ...................................................................... 169
5.1
5.1.1
Basic architecture of the CBRP .................................................................................. 170
5.1.2
Main changes of the CBRP from 2001 to 2010 ..................................................... 171
Analysis of policy drivers ................................................................................................. 181
5.2
5.2.1
Climate change ................................................................................................................ 182
5.2.2
Supporting the construction industry ........................................................................ 195
vi
Contents
5.2.3
Recession ........................................................................................................................... 199
5.2.4
Budgetary considerations ............................................................................................. 202
5.2.5
Change of personnel....................................................................................................... 210
Conclusion ........................................................................................................................................... 212
6
Evaluation........................................................................................................................................ 214
6.1
Comparison of official results .......................................................................................... 216
6.2
Evaluating energy efficiency programmes ................................................................... 220
6.2.1
Rebound effects ............................................................................................................... 220
6.2.2
Prebound effects .............................................................................................................. 222
6.2.3
Free rider effects ............................................................................................................. 223
Evaluations of the Supplier Obligation ......................................................................... 224
6.3
6.3.1
In-depth appraisal of the evaluation .......................................................................... 225
6.3.2
Prebound and rebound effects ..................................................................................... 226
6.3.3
Free rider effects ............................................................................................................. 229
Evaluations of the CO2 Building Rehabilitation Programme ................................. 230
6.4
6.4.1
In-depth appraisal of the evaluations ........................................................................ 231
6.4.2
Prebound and rebound effects ..................................................................................... 232
6.4.3
Free rider effects ............................................................................................................. 233
Conclusions ......................................................................................................................................... 236
7
Cross-case analysis: politics of change ................................................................................... 238
Mechanisms of policy change .......................................................................................... 238
7.1
7.1.1
External pressures ........................................................................................................... 239
7.1.2
Friction ............................................................................................................................... 243
7.1.3
Policy feedback ............................................................................................................... 249
7.1.4
Summary ........................................................................................................................... 255
Hypotheses ............................................................................................................................ 256
7.2
7.2.1
Hypothesis 1 ..................................................................................................................... 256
7.2.2
Hypothesis 2 ..................................................................................................................... 259
7.2.3
Hypothesis 3 ..................................................................................................................... 259
7.2.4
Hypothesis 4 ..................................................................................................................... 261
7.3
8
Limitations of the analysis ................................................................................................ 262
Conclusions ..................................................................................................................................... 264
8.1
Answering the research questions................................................................................... 264
vii
Contents
9
8.2
Contribution .......................................................................................................................... 268
8.3
Policy implications .............................................................................................................. 269
8.4
Avenues of further research .............................................................................................. 273
Bibliography ................................................................................................................................... 275
viii
Contents
Figures
Figure 1: Organisation of the thesis according to the main objectives and the stages of
systematic process analysis ........................................................................................... 9
Figure 2: Conceptual framework ......................................................................................... 42
Figure 3: Mechanism of the Supplier Obligation in Great Britain ...................................... 83
Figure 4: Annual energy savings target of SO in TWh ....................................................... 87
Figure 5: Mix of measures in different SO schemes ........................................................... 98
Figure 6: Fuel price index numbers relative to the GDP deflator (2005=100) ................. 122
Figure 7: Fuel price indices in the domestic sector in real terms Q1 2000 to Q1 2010 .... 124
Figure 8: Newspaper coverage of climate change in the UK ............................................ 134
Figure 9: Number of households in fuel poverty in England (1996-2009) ....................... 147
Figure 10: Uptake of insulation measures under the SO (1994-2010) .............................. 156
Figure 11: KfW schemes and financial arrangement ........................................................ 171
Figure 12: Federal funding of the CBRP KfW definition & loans and grants issued ....... 173
Figure 13: Newspaper coverage of climate change in Germany ....................................... 187
Figure 14: Turnover and employment in the German construction industry 1995-2010 .. 196
Figure 15: Ifo Business Climate Index in Germany .......................................................... 200
Figure 16: Public debt in Germany as a proportion of GDP ............................................. 203
Figure 17: CO2-emissions in Mt (lifetime) and TWh (lifetime) saved per year ............... 218
Figure 18: Financial support for refurbishment and impact on decisions of households .. 235
ix
Contents
Tables
Table 1: Positive and negative feedback in Punctuated Equilibrium Theory and historical
institutionalism ............................................................................................................ 40
Table 2: Taxonomy of policy components with examples .................................................. 61
Table 3: Main elements of policy change in case of the SO ............................................... 84
Table 4: History of the SO in Great Britain ........................................................................ 85
Table 5: Main policy drivers of the SO ............................................................................. 102
Table 6: KfW housing refurbishment programmes ........................................................... 170
Table 7: Main elements of policy change in case of the CBRP ........................................ 172
Table 8: CBRP loan requirements for packages of measures and refurbishment according
to new built standard or better from 08/2011-03/2009 .............................................. 178
Table 9: Main policy drivers of the CBRP ........................................................................ 181
Table 10: Total expenditure, carbon and energy savings 2002-2007* .............................. 219
Table 11: Reduction factors applied to cavity wall insulation in various periods of the SO
................................................................................................................................... 229
Table 12: Summary of policy change mechanisms identified ........................................... 255
x
Abbreviations
Abbreviations
ACF: Advocacy Coalition Framework
BEEF: British Energy Efficiency Federation
BERR: Department for Business, Enterprise and Regulatory Reform
BMU: Bundesministeriums für Umwelt, Naturschutz und
[Bundesministerium für Umwelt, Naturschutz und Reaktorsicherheit]
Reaktorsicherheit
BMWi: Bundesministerium für Wirtschaft und Technologie [Federal Ministry of
Economics and Technology]
BMVBS: Bundesministerium für Verkehr, Bau und Stadtentwicklung [Federal Ministry of
Transport, Building and Urban Development]
CBRP: CO2 Buildings Rehabilitation Programme
CDU: Christlich Demokratische Union Deutschlands [Christian Democratic Union of
Germany]
CESP: Community Energy Savings Programme
CERT: Carbon Emissions Reduction Target
CFLs: compact fluorescent lamps
CHP: combined heat and power
CO2: carbon dioxide
CSU: Christlich Soziale Union [Christian Social Union]
DECC: Department of Energy and Climate Change
DEFRA: Department for Environment, Food and Rural Affairs
DENA: Deutsche Energieagentur [German Energy Agency]
DETR: Department of the Environment, Transport and the Regions
DTI: Department of Trade and Industry
ECO: Energy Company Obligation
EEC: Energy Efficiency Commitment
EESoP: Energy Efficiency Standards of Performance
EPCs: Energy Performance Certificates
ERA: Energy Retail Association
EST: Energy Saving Trust
FDP: Freie Demokratische Partei [Free Democratic Party]
FPAG: Fuel Poverty Advisory Group
GHG: greenhouse gas emissions
HECA: Home Energy Conservation Act
IEA: International Energy Agency
xi
Abbreviations
IPCC: Intergovernmental Panel on Climate Change
KfW: Kreditanstalt für Wiederaufbau [German Development Bank]
kWh: Kilowatt hour
LCP: Least-Cost-Planning
mt: million metric tonnes
MS: Multiple Streams
MWh: Megawatt hour
NEA: National Energy Action
NIA: National Insulation Association
NGO: Non-governmental Organization
OFFER: Office of Electricity Regulation
OFGAS: Office of Gas Supply
OFGEM: Office of Gas and Electricity Markets
PET: Punctuated Equilibrium Theory
PES: Public Electricity Suppliers
PG: Priority Group
PIU: Performance and Innovation Unit
ppm: parts per million
RECs: Regional Electricity Companies
RO: Renewables Obligation
SO: Supplier Obligation
SPD: Sozialdemokratische Partei Deutschlands [Social Democratic Party of Germany]
TWh: Terawatt hour
UK: United Kingdom
US: United States of America
xii
Introduction
1 Introduction
Practical politics consists in ignoring facts.
Henry Adams (1918)
The above quote by the American historian and novelist Henry Adams suggests that
politics is not driven primarily by ‘objective’ information and ‘facts’, but, instead, is a
messy and rather unpredictable process. However, the aspect of politics is often treated
lightly in energy studies – a field predominantly composed of research trying to identify
facts including technological questions, the cost of different technologies, the economics
of policy instruments, and their distributional impacts. This is no different for the academic
literature on energy efficiency policy – key points of interest regard the technical potential
of energy efficiency measures, the economic aspects of energy efficiency, and the policy
apparatus that should be adopted theoretically.
A lot of work can be classified as analysis for rather than of policies. Very few studies,
though there are some exceptions, analysed actual policies from a public policy
perspective. Although it is useful to debate the effectiveness of existing policies and
appraise the theoretically possible range of various instruments, it is equally important to
understand why we end up with some policies and not with others, what the determinants
of policy change and stagnation are, and whether policy change follows certain patterns
and is driven by specific causal mechanisms. Such an investigation can help explain why
despite factual evidence policy change takes its own direction, often to the frustration of
the expert.
1
Introduction
This raises the question what, if not ‘facts’, does determine policy change. Politics is all
about change. Thus, in order to understand change, one has to comprehend the significance
of politics. A better understanding of past policy trajectories and the political drivers is
crucial for the future design of effective long-term policies. Alas, in the past, energy
efficiency policies were often designed for rather short time horizons lacking a long-term
framework. Considering the magnitude of change required, there is evidence that a
structural transformation will be necessary in order to achieve long lasting reductions of
energy consumption and their associated carbon emissions (Boardman 2012; Ekins 2004;
Eyre 1997; Skea 2012; Unruh 2000; Verbong and Geels 2007). Short-term policy design is
ill suited to achieve such a transformation and more substantial and long-term policies are
required that focus not only on technological measures, but also explicitly consider the
politics associated with particular instruments. Therefore, the rationale behind this
investigation is to understand why we ended up with those policies in the first place and
what kind of mechanisms caused the changes over time.
1.1 Subject of investigation
This thesis analyses the dynamics of domestic energy efficiency policy in the UK and
Germany where ambitious policies have been implemented over the last decades. It asks
the question:
How and why did domestic energy efficiency policy change?
By default, the analysis required to answer this question needs to focus on depth rather
than breadth. There are numerous energy efficiency policies in place and it would be a
herculean task to analyse the politics of change for all of them in one doctoral thesis.
Instead, this thesis focuses on two policy instruments only, the German CO2-Building
Rehabilitation Programme (CBRP) and energy savings obligations in Britain.
2
Introduction
The CBRP is a large loan and grant scheme for financing energy efficiency
refurbishments. It is funded by the tax payer and administered by the Kreditanstalt für
Wiederaufbau (KfW), a publically owned bank with a focus on providing finance for
infrastructure projects in Germany and developing countries. Energy savings obligations in
Britain oblige large energy suppliers to implement energy efficiency measures in homes
and is funded by energy bills.
The rationale for choosing these two policy instruments is as follows: First, both
instruments are established and have been in place for some time, a requirement for
observing long-term changes. Second, they constitute the key policy instrument in the
policy area of energy efficiency and thereby represent the overall policy approach in the
respective countries well. A detailed justification for case selection is provided in Chapter
3.
1.2 Theoretical approach: a brief introduction
There is a rich literature on the theories of policy change, a lot of which evolved over the
last two decades. Three theories of policy change are particularly prominent: the Advocacy
Coalition Framework (ACF), the Multiple Streams approach (MS), and the Punctuated
Equilibrium Theory (PET) (Capano 2009; John 1998, 2003). All three theories are a
response to incrementalism, a theory that dominated public policy analysis in the 1960s
and 1970s and predicted that most policy changes would be relatively minor. Chapter 2
discusses these theories in detail, but it is worth mentioning that the three mainstream
theories of policy change, ACF, MS, and PET, emphasise short-term crisis-type events as
the main drivers for significant policy change. This is at odds with the observation that
many policy changes seem to be driven by more gradual, long-term pressures opposed to
short-term events.
3
Introduction
More recently, more comprehensive theories of gradual pressures and policy change have
emerged. This thesis will build on these theories and apply them to the policy area of
energy efficiency. These theories include the concept of friction developed by scholars of
PET (Baumgartner 2011; Baumgartner et al. 2009; Baumgartner and Jones 2009; Jones
and Baumgartner 2005, 2012) and the policy feedback literature (Béland 2010; Pierson
1993a, 2004b; Skocpol 1992; Weaver 2010). Together, the theories can help explain how
pressures accumulate and eventually result in policy change (friction), but also how
existing policies have slowly developing consequences affecting the policies themselves
(policy feedback). Given their suitability to explain more gradual and long-term change,
the thesis builds on the concept of friction and the policy feedback literature in order to
approach the research question conceptually.
1.3 Importance of the domestic sector for climate change policy
Studies indicate that buildings alone contribute more than 30% to energy-related global
CO2 emissions (GEA 2012), which is similar to what the entire industrial sector emits
(IPCC 2007; Stern 2007). A large share of this comes from residential buildings, which are
responsible for almost one quarter of global energy demand and are the most important
contributing segment within the building sector (GEA 2012). This means that any attempts
to reduce GHG emissions in industrialised countries must include measures to reduce
emissions from residential buildings, particularly because the potential is substantial
(Boardman 2007, 2012; Ekins et al. 2002). According to the IPCC, energy efficiency in the
building sector plays a key role for climate change mitigation policy and can deliver
significant reductions of GHG emissions. The IPCC estimates that of the projected
baseline emissions at least 29% can be reduced cost-effectively in the residential and
commercial sectors by 2020, the highest reduction potential among all sectors studied in
the latest IPCC mitigation report (IPCC 2007).
4
Introduction
Even though historical trends in Germany (AG Energiebilanzen 2012; DIW and ISI 1997)
and the UK (DECC 2012c) show that all energy use in domestic buildings increased
substantially since the 1970s, in more recent years energy consumption has fallen
significantly in both countries. In Germany, households consumed 15% less compared to
2000. For heating oils, the reduction is even higher and amounts to 22% (Statistisches
Bundesamt 2012b). The UK shows similar trends – in 2011 households consumed 17%
less energy compared to 2000 with gas consumption dropping by 20% (DECC 2012d).
Domestic electricity consumption did increase in both countries though. The observed
reduction of energy demand for household heating clearly is the result of energy efficiency
policies (Centre for Economics and Business Research 2011; Schlomann and Eichhammer
2012).
1.4 Research questions and aim
The aim of this research is twofold: First, to understand the conditions under which energy
efficiency policies have changed in the past to inform future policy development and help
more realistic planning of those. Second, to contribute to the policy change literature
putting the emphasis on gradual pressures, including feedback mechanisms, and their
impact on policy instruments. More specifically, the key objectives of this research are:

conceptualising the role of long-term pressures in policy change;

describing and understanding the different kinds of long-term pressures in two case
studies (the SO and the CBRP); and

comparing and analysing the causal mechanisms of long-term pressures.
Following the above, this thesis asks a number of more specific research questions:
1) How did the key domestic energy efficiency policy instrument in the UK and
Germany change since their inception?
5
Introduction
2) What kind of pressures were the most important drivers of policy change?
3) How did the principal policy instrument of the policy subsystem of domestic
energy efficiency policy respond to gradual pressures?
4) How well can the concept of friction explain those processes? If there is evidence
for friction, which factors caused friction?
5) What role did policy feedback processes play in policy change?
1.5 Approach
The focus of the analysis will lie on the policy subsystem of domestic energy efficiency
policy. A policy subsystem ‘consists of actors from a variety of public and private
organizations who are actively concerned with a policy problem or issue […] and who
regularly seek to influence public policy in that domain’ (Sabatier 1998, p. 99) The
challenge of this thesis is to understand what drives change in the policy subsystem of
domestic energy efficiency policy focusing on the two policy instruments mentioned
before. Rather than simply determining the causes of policy change it has been argued that
‘research could identify more precisely the causal mechanisms driving policy change [my
emphasis]’ (John 2003, p. 481). Causal mechanisms refer to the process by which an
independent variable impacts on a dependent variable. For example, an energy crisis is
often cited as a cause of policy change. In order to explain why an energy crisis actually
drives policy change, i.e. the causal mechanism, one needs to understand the underlying
processes that connect the crisis with the outcomes of change such as the response of
actors to the crisis and the way those actors make decisions that eventually lead to policy
change. Hence, causal mechanisms encompass the causal chain that leads to an outcome.
6
Introduction
Methodologically, an approach known as systematic process analysis (also called process
tracing) (Hall 2003, 2008) is followed. Systematic process analysis involves the following
steps:
1. Theory formation: approach each case with a range of theories and potential causes
to test one theory against another;
2. Deriving predictions: develop predictions that are consistent with one theory but
inconsistent with its principal rivals;
3. Making observations: observation of correlation as in conventional comparative
method but also of the processes whereby an outcome is caused; and
4. Drawing conclusions: assessment of the extent to which the theories explain the
outcome.
The thesis is organised along these four steps as outlined in the next section.
1.6 Organisation of the thesis
On the basis of the four steps of systematic process analysis, the thesis contains the
following chapters:

Chapter 2 Theories of policy change: friction and policy feedback: Chapter 2
introduces the theoretical underpinning of this thesis. It starts with a critique of existing
theories of policy change and moves on to developing a conceptual framework based
on the policy feedback literature and the concept of friction coming out of Punctuated
Equilibrium Theory.

Chapter 3 Methodology: Chapter 3 provides the methodology used in this thesis. It
explains the research design chosen, the rationale for opting for a case study approach,
and justifies the case selection. Furthermore, the chapter spells out the data sources and
7
Introduction
research methods used in detail including their limitations. Finally, the chapter presents
how the analytical framework is operationalised.

Chapter 4 and Chapter 5: The empirical part of this thesis is located in chapters 4
and 5 and consists of two cases. Chapter 4 covers the first case, energy savings
obligations in the UK, whereas Chapter 5 focuses on the German CO2-Building
Rehabilitation Programme. Chapter 4 is longer than Chapter 5 for the simple reason
that the UK case covers a longer time span.

Chapter 6 Effectiveness of the two policies: This chapter is dedicated to the question
of effectiveness i.e. which policy instrument is ‘better’ in terms of reducing carbon and
saving energy.

Chapter 7 Mechanisms of long-term policy change: cross-case analysis: This
chapter links the empirical research to the theoretical presumptions and discusses how
the processes of policy change identified in the two cases compare with the
expectations based on the theory presented in Chapter 2. Furthermore, the chapter
derives a more general model of long-term policy change based on the results of the
analysis.

Chapter 8 Conclusions: The final chapter summarises the main argument, answers the
research questions, and reflects on the limitations of the analysis. Avenues for further
research and policy implications are discussed.
The chapter structure is aligned with both the key objectives and the stages of systematic
process analysis. Figure 1 illustrates how each chapter responds to the objectives and
relates to the four stages of systematic process analysis.
8
Introduction
Figure 1: Organisation of the thesis according to the main objectives and the stages of systematic
process analysis
9
Theories of policy change
2 Theories of policy change: friction and policy feedback
He who loves practice without theory is like the sailor who boards ship without a rudder and
compass and never knows where he may cast.
Leonardo da Vinci (quoted in Kline 1990, p. 224)
A great deal has been written about the causes of policy change. Theories trying to
conceptualise these causes became more and more sophisticated, covered an increasing
number of issues, were tested both quantitatively and qualitatively, and subsequently
improved their robustness. Therefore, any inquiry into the causes of policy change is
well advised to start by considering what these theories have to say about the drivers of
policy change. Hence, this research project started with a review of the literature on
policy change in search for potential theories that might help guide and interpret the
case study results. At the same time, suitable case studies were identified that would
provide an interesting and promising object of investigation. While the choice of cases
was challenging, the decision for or against various theories was a major hurdle: Initial
analysis of both case studies showed that the policy instruments investigated changed
substantially over the years. And both cases indicated that it was long-term pressures
such as the perceived impact of climate change and effects of the policy itself that
fuelled the changes observed. However, the initial review of the theories of policy
change suggested that substantial change was generally conceptualised as the result of
short-term, exogenous pressures such as crises. In the absence of such events, only
10
Theories of policy change
incremental change appeared to be likely.1 Advocates of the various theories of policy
change often point out precisely this weakness of many theories as will be shown
below. Fortunately, there are now first theoretical approaches that conceive of long-term
pressures as important mechanisms of policy change. This thesis utilises and expands
these concepts to make sense of the case study results at a more general level and
contribute to the theoretical discussions around the politics of long-term change.
This chapter is comprised of four sections. First, it introduces and critically discusses
the most prominent theories of policy change. Second, while the strengths of the
theories discussed are acknowledged, they are also criticised for overemphasising the
role of external events and underplaying the importance of long-term pressures. Third,
in search of promising theoretical approaches that relate to long-term policy drivers, the
chapter introduces the concept of friction and draws on the literature around policy
feedback. Finally, the conceptual framework of this thesis is presented combining both
the friction concept and the policy feedback literature. The operationalisation of the
conceptual framework and its application to the case studies (including the methodology
chosen) is presented in a separate chapter before the case study analysis is carried out.
2.1 Theories of policy change
There are various theories about policy change, each with different perspectives on the
policy process emphasising particular elements or mechanisms. Hence, no single theory
can explain policy change in all circumstances and some theories have more
explanatory power concerning certain aspects of policy change than others. All of these
theories can be used to shed light on the causal mechanisms of policy change. Although
1
This is, of course, a stylisation of the literature and the theories are much more complex. However,
the overall picture can be characterised in such a two-dimensional way.
11
Theories of policy change
different theories of policy change make diverse assumptions on the causal
mechanisms, they share an implicit critique of the idea that public policy follows an
ideal ‘policy cycle’ (Lasswell 1971), also called the stages heuristic (Sabatier 2007b),
and try to explain how political processes are rather than how they should be. While the
policy cycle has been criticised for its lack of causal explanation, it is a useful analytical
map of the policy process and all of the theories refer to the importance of different the
stages in the policy process.
The list of theories focusing on policy change is long and one can easily get lost in the
multiple facets of the different concepts. Here, initially only the three most prominent
theories are analysed with regard to their conceptualisation of the causes of policy
change. These are the Advocacy Coalition Framework (ACF), the Multiple Streams
approach (MS), and the Punctuated Equilibrium Theory (PET) (Capano 2009; John
1998, 2003). It is, however, useful to include the theoretical starting point of these
theories in the review because a substantial component of the theories discussed is still
based on an early theory of policy change, namely incrementalism. Starting with
incrementalism, each theory will be introduced followed by a short review of its
limitations particularly with regard to accounting for long-term pressures.
2.1.1 Incrementalism
Incrementalism is the basis on which most, if not all, of the theories of policy change
rest and it is important to acknowledge its lasting influence on today’s theoretical
accounts. The incrementalist approach is mainly associated with Charles Lindblom’s
work (Dahl and Lindblom 1953; Lindblom 1959, 1979) who argued that the rational
policy model in the sense of policy makers assessing the problems, finding suitable
solutions, and then implementing them, did not stack up in the real world of policy
12
Theories of policy change
making. He wanted to separate the ‘ought from the is [my emphasis]’ (Lindblom 1979,
p. 517).
At the basis of incrementalism lies the assumption that policy change follows a logic of
‘muddling through’, and that decision makers cannot make rational decisions due to the
complexity of all the different alternatives and their bounded rationality, a concept first
developed by Herbert Simon (1959). Hence, decision makers tend to choose among the
alternatives already at hand and make only slight adjustments. As a result, political
change happens slowly due to a status quo bias and the constraints under which decision
makers make their decisions (Lindblom 1959). In sum, incrementalism suggests that
incremental change is the norm whereas more radical change is rather unlikely.
For many decades, incrementalism was the dominant theory of policy change (Howlett
and Migone 2011) and formed the basis of subsequent theoretical approaches
conceptualising policy change. Incrementalism attracted many supporters but of course
also had its critics. Assessing the criticism voiced by other scholars, Weiss and
Woodhouse (1992) summarise the key points raised and stress that most critics accuse
incrementalism of being a) insufficiently proactive, goal oriented, and ambitious; b)
excessively conservative; c) only useful in too limited a range of decision contexts; and
d) too hostile to analysis. A further problem of incrementalism is that it is not clearly
defined – what magnitude of change over which time span can be classified as
‘incremental’ and at what point does it exceed incremental change? The ranges used
varied considerably, for example when investigating budgetary changes, scholars used
definitions for changes expected under incrementalism from 2% to 30% (Anderson and
Harbridge 2010).
Probably the point that advanced theory development most was the critique that
incrementalism is too conservative – this triggered the search for alternative theories
13
Theories of policy change
that could explain both incremental and radical change. It is probably fair to say that all
of the major theories of policy change appreciated the contribution of incrementalism to
the literature, but also tried to develop more comprehensive accounts of policy change
based on incrementalism.
2.1.2 Advocacy Coalition Framework (ACF)
ACF is an influential theoretical concept used by scholars studying policy change and
there are a large number of studies employing ACF (Weible et al. 2009). The main idea
of the ACF is that the actors of a given policy subsystem (for example domestic energy
efficiency policy) form advocacy coalitions consisting of stakeholders who share the
same normative and causal beliefs related to policy objectives (Sabatier 1988; Sabatier
and Jenkins-Smith 1993). Those beliefs are classified into deep core beliefs, policy core
beliefs, and secondary aspects.
Deep core beliefs refer to fundamental normative and ontological axioms (e.g. belief in
free markets). Deep core beliefs are extremely difficult to change and Sabatier (1988)
compares a change in those beliefs to religious conversion.
Policy core beliefs are defined as the position concerning the strategies for achieving the
goals of the deep core (e.g. belief in deregulation of energy markets). The policy core
beliefs are difficult to change, but change can happen as a result of external pressures
such as major socio-economic change, crises such as wars and energy price shocks (e.g.
the 1973 oil crisis), system-wide change due to electoral shifts (e.g. election of Ronald
Reagan and Margret Thatcher) but also subsystem spillovers (i.e. when policy change in
another sector impacts on the policy subsystem in question).
Secondary aspects cover narrower beliefs about specific aspects of the problem and
policy implementation (e.g. in favour of energy efficiency building standards and
14
Theories of policy change
opposed to white certificate trading schemes). Secondary aspects change fairly easily in
a process of policy learning as a result of growing experience, new information, or if
their alteration can help to protect the policy core beliefs from criticism.
Whereas earlier versions of the ACF focused only on two forces of policy change,
namely external subsystem events and policy learning, more recent versions introduced
two additional paths of policy change (Weible et al. 2009). A third force of change is
internal subsystem events due to failure of the current policies. The fourth path to policy
change refers to negotiated agreements between advocacy coalitions. Within this
approach nine conditions are presented as determining the likelihood of policy change: a
hurting stalemate, effective leadership, consensus-based decision rules, diverse funding,
duration of process and commitment of members, a focus on empirical issues, an
emphasis on building trust, and lack of alternative venues (Sabatier and Weible 2007).
The ACF is a comprehensive framework to investigate policy change. But there has, of
course, also been some criticism of the ACF: In the ACF, external perturbations and
policy learning are the two main phenomena that lead to policy change. However,
policy learning and external perturbations do not always lead to change (Mintrom
1996). Sabatier and Jenkins-Smith (1999) responded to such criticism by adjusting their
framework to conceptualise external perturbations as necessary, but not sufficient
causes of policy change. Furthermore, they introduced internal shocks and negotiated
agreements between coalitions as factors influencing policy change (Sabatier and
Weible 2007). Nevertheless, in their work some causal mechanisms remain
underspecified, particularly concerning the role of policy entrepreneurs (Mintrom
1996). Another point of critique is that ACF scholars mainly focus on how beliefs
change, but not how policy decisions are made (Schlager 1999). Also, it is not clear how
15
Theories of policy change
beliefs lead to collective action of an advocacy coalition and what strategies coalitions
are pursuing to press for political change (Schlager 1995).
2.1.3 Multiple streams (MS)
Another influential theory used by scholars investigating policy change is the MS
approach. Drawing on the Garbage Can Model by Cohen et al. (1972), MS
conceptualises the policy process as three unrelated streams: problems, policies, and
politics (Kingdon 1984, 1995, 2002; Zahariadis 1992, 1995, 1996, 1999).
The problem stream refers to issues of concern. Only if an issue is perceived as a
problem, will actors look for solutions and alternatives. There are various ways by
which issues can become a problem. Actors monitor data and look for potential
problems such as the number of car accidents, the rate of global warming, budget
deficits and the like. Focusing events such as wars and natural catastrophes can draw
attention to a particular problem, as can feedback from existing policies (Zahariadis
1999). However, whether a problem finds recognition also depends on how it is framed
(Kingdon 2002), and on the beliefs and values of the people interpreting it (Zahariadis
1999).
The policies stream is conceptualised as a ‘primeval soup’ where ideas float around
bump into one another, combine and recombine in different ways. The concept
originated in biology and the idea of natural selection. Kingdon (1984) argues that
various actors try out different ideas and that some of these ideas survive the process of
selection whereas others fail to do so. An idea must mature before adoption and be
‘ripe’ for the policy process. Hence only certain ideas will conform to the values of
policy makers and find recognition (Zahariadis 1999).
16
Theories of policy change
The politics stream is composed of three elements: national mood, pressure group
campaigns, and administrative or legislative turnover (Kingdon 2002; Zahariadis 1999).
Each of these elements can prevent or help an idea to find recognition in policy making.
The national mood remains somewhat underspecified and does not only refer to public
opinion as measured by opinion polls (Kingdon 1984) but ‘the notion that a fairly large
number of individuals in a given country tend to think along common lines and that the
mood swings from time to time’ (Zahariadis 1999, p. 76). Pressure group campaigns
can also affect whether policy makers support or disregard ideas of policy change. For
example, if there is widespread interest group support of market-based instruments in
the area of energy efficiency policy makers are more likely to pursue such ideas and
vice versa. Change in government as happens during ‘critical elections’ (Brady 1988;
Burnham 1970; Key 1955) can also have a major impact on the likelihood that an idea
is adopted. A good example is the election of Ronald Reagan in 1980; Reagan
subsequently blocked attempts at introducing energy efficiency standards in the US that
had been supported by his predecessor (Varone and Aebischer 2001). In order to be able
to operationalise the politics stream, Zahariadis (1999) proposes to merge the three
elements national mood, pressure group campaigns, and administrative or legislative
turnover into the single variable ‘ideology of governing parties’. Zahariadis argues that
such amalgamation is justified because parties dominate the politics stream.
Policy change is more likely to happen when all three streams are joined together and a
‘policy window’ or ‘window of opportunity’ is created. Those windows are situations in
which a ‘problem is recognised, a solution is developed and available in the policy
community, a political change makes the right time for policy change, and potential
constraints are not severe’ (Kingdon 1984, p. 174). Policy entrepreneurs can take
advantage of those policy windows by ‘hooking their solutions to problems or by
17
Theories of policy change
ensuring that proposals from the policy stream are considered when the political
conditions are right’ (Kingdon 2002, p. 102). Kingdon compares this process to a surfer
waiting for the big wave. The surfer has to be ready to paddle when it comes, and if he
isn’t he is not going to ride the wave.
While being very popular amongst scholars dealing with policy change, MS also
received some criticism. One of the problems of MS was that it concentrated mainly on
agendas and not on policy implementation and outcomes (Capano 2009; John 1998).
While providing a compelling approach for explaining agenda setting, Kingdon (1984)
did not provide a framework to understand the impact of agenda setting on policy
implementation. There have been attempts to extend MS to the decision-making process
(Zahariadis 1995, 1996) and even the whole of the policy process (Saint-Germain
1996). In a revised version of MS, Kingdon (1995) extended the framework to the entire
policy cycle. Drawing on ACF, Kingdon’s model has been further developed by Ness
(2010), who added two additional elements namely the policy milieu and the policy
field, representing the governance structures and the policy environment in which
decisions are made. A focal point of criticism is also the interplay of the three streams.
As shown above, MS conceptualises three streams independent from each other.
However, this has been criticised for ignoring the fact that the three streams are actually
interdependent and linked to each other (Capano 2009; Mucciaroni 1992). Finally, MS
has been attacked for not really being a theory, but rather a heuristic device that can be
used in policy analysis (King 1985). For the reasons cited above, it has been argued that
scholars found it difficult to operationalise MS and, in contrast to PET and ACF, few
empirical studies were carried out (Sabatier 2007a), although Nowlin (2011) identified
at least three studies published in 2009 and 2010 that successfully tested hypotheses
based on MS. However, MS is probably the least applied of the three main theories of
18
Theories of policy change
policy change – both PET and ACF were used in numerous studies and continue to be
used by many researchers.
2.1.4 Punctuated Equilibrium Theory (PET)
The PET model is borrowed from evolutionary biology (Eldredge and Gould 1972) and
assumes that the political system (like biological systems) reaches equilibrium, a time of
stability, which is then ruptured by exogenous events (Baumgartner and Jones 1991,
1993, 2002a; John 1998, 2003; True et al. 2007b). Radical policy change happens at
certain critical points. As a result of the disruption a new equilibrium is reached which
remains more or less stable until it is ruptured again. During times of stabilisation,
processes such as positive feedback keep the system in equilibrium. Baumgartner and
Jones (1993) introduce the concepts of policy image and policy venue as central
explanations for the punctuated equilibrium model:
A policy image refers to ‘how a policy is understood and discussed’ by the actors
involved in the policy process (Baumgartner and Jonesp. 25). If the policy image is
altered significantly, the system is ripe for punctuation. PET differentiates between
those images favourable and those that are detrimental to proponents of a given policy.
In the case of domestic energy efficiency policy, positive images of policy measures
might be energy savings, cost reduction, carbon mitigation etc. Negative images can
include administrative burden, rebound effect, interfering with consumption habits etc.
Public attention can focus on negative or positive images for long periods of time, but
ultimately may shift at some point (Baumgartner and Jones 1991). Proponents of a
given policy tend to focus on one set of images in favour of their proposals whereas
opponents are likely to refer to another set of images discrediting those proposals.
Policy makers are inclined to try to establish policy monopolies by ensuring the
dominance of those policy images that support their policies and reject alternatives
19
Theories of policy change
(Baumgartner and Jones 2009). It follows that there is a constant struggle about policy
images affecting the policy process in which actors try to champion or mould particular
policy images that are in line with their proposals.
Policy venues are conceptualised as the ‘existing set of political institutions’ or ‘the
venues of policy action’ (Baumgartner and Jones 1991, p. 1045). Hence venues are the
‘institutional locations where authoritative decisions are made concerning a given issue’
(Baumgartner and Jones 2009, p. 32). Some policy images find a favourable reception
in certain policy venues but not in others. For example, the image of energy efficiency
as a means to reduce carbon emissions might not be considered by a government
department mainly interested in economic efficiency, but another agency with a focus
on environmental protection might be very supportive. Consequently, actors will try to
find a policy venue that is receptive to the policy image they want to foster by what is
called ‘venue shopping’ (Pralle 2003). Furthermore, actors pursue image manipulation
in order to find recognition with their ideas (Baumgartner and Jones 1991). For
example, framing energy efficiency as a cost-saving opportunity rather than a means to
protect the environment might draw support from venues that previously would not
consider energy efficiency activities. Hence policy images are linked with policy venues
in the sense that some venues are more receptive to certain images than others and vice
versa. However, policy venues and decision-making structures themselves may also
change over time with significant impact on policy outcomes (Baumgartner and Jones
2009). For example, in the UK energy policy was mainly the responsibility of the
Department of Business, Enterprise and Regulatory Reform (BERR) until the creation
of the Department of Energy and Climate Change (DECC) in October 2008. DECC
brings together both energy policy and climate change policy which was previously the
responsibility of the Department for Environment, Food and Rural Affairs (DEFRA). It
20
Theories of policy change
is very likely that the coupling of the two issues in one department will have a
significant impact on policy decisions.
As a result, venues and images reinforce themselves: When the policy images change,
venue change becomes more likely and when policy venues change, image change is
more likely to happen. Such feedback mechanisms can lead to ‘the rapid creation,
destruction, or alteration of policy subsystems’ (Baumgartner and Jones 1991, p. 1045).
This can happen when a change outside of the subsystem impacts on either policy
image or venue (for example a major crisis such as the accident at the Fukushima
nuclear power station in 2011 which radically transformed the image of nuclear power
in Germany across the spectrum of political parties). Negative feedback, as defined by
PET, leads to stability and incrementalism in situations ‘where pressures from one side
lead to counter-pressures from another side, and in general where self-corrective
mechanisms keep the system on an even keel’ (Baumgartner and Jones 2002b, p. 6).
Negative feedback in PET can be compared to a thermostat adjusting to outside
temperatures, it counter-balances external pressures. An example of negative feedback
is policies aimed at controlling inflation in order to keep the economy stable. The
pattern of policy change thus follows long periods of incremental change ruptured by
external pressures leading to short-term (but long lasting) policy change.
PET has been criticised for suggesting that major change always needs to come from
outside of the policy subsystem in the form of exogenous shocks leading to punctuation
(Lieberman 2002; Streeck and Thelen 2005). Even if external pressures create new
policy venues and images, policy change does not always take place. Actors might
block policy change with the result that it never happens or gets delayed (Walgrave and
Varone 2008). However, recent revisions of PET (‘second generation PET’) added
mechanisms such as the notion of friction that account for policy change driven by
21
Theories of policy change
long-term pressures and the build-up of these over time (Jones and Baumgartner 2005).
Therefore, in second generation PET it does not have to be a major exogenous event,
but can equally be ‘relatively minor events that add up over longer periods of time’
(True et al. 2007b, p. 160). The concept of friction will be explained in great detail at a
later point in this thesis because it is one of the two building blocks of the theoretical
framework of this thesis. Furthermore, PET-based studies have been criticised for
concentrating on the quantitative aspects of policy change and neglecting the qualitative
elements (Howlett 1997). PET was first developed in the US and applied to American
politics, which is why the majority of studies focussed on policy change in the US. It is
questionable to what extent PET can be applied in its original form to other political
systems. However, there are now serious efforts to apply PET to countries other than
the US, mainly facilitated by the Comparative Agendas Project (John 2006), and
applications of PET in Europe in recent years indicate that it can also be applied
successfully outside the US. Another point of critique is that PET research concentrates
on agendas and on the tone of political debates. However, when looking at policy
outcomes there is less evidence in support of PET (Givel 2010).
2.2 General critique
The four theories (ACF, MS, PET, incrementalism) make a substantial contribution to
the understanding of the dynamics of public policy and provide sophisticated
frameworks for policy analysis. There is a certain degree of overlap among the theories
and many scholars argue that ACF, MS, and PET are complementary (Cairney 2009;
John 2003; Meijerink 2005; Zahariadis 1998), although the focus of each theory lies on
different elements of the policy process and their underlying conceptualisation of the
mechanisms of change is not always the same. One important similarity is that all of the
22
Theories of policy change
theories of policy change except incrementalism rely heavily on exogenous events as
the key driver of major policy change (Capano 2009; Wood 2006):
ACF hypothesises that ‘significant perturbations external to the subsystem are a
necessary, but not sufficient, cause of change in the policy core attributes of a
governmental program’ (Sabatier and Weible 2007, p. 220). Hence the main driver of
significant policy change is exogenous events such as major socioeconomic changes,
public opinion shifts, changes in the system-governing coalition, and policy decisions
and impacts from other subsystems (Cairney 2007; Lowry 2006; Nohrstedt 2005;
Nowlin 2011; Schlager 1999; Smith 2000). As already mentioned, more recent versions
of the ACF, however, do also account for internal shocks (within the subsystem) and
negotiations between advocacy coalitions as additional mechanisms for change
(Sabatier and Weible 2007). So far, these factors remain underplayed in the ACF
literature.
Also depending to a large extent on external events, MS refers to ‘focusing events’ such
as crises and critical elections as one of the key reasons for the opening of windows of
opportunity, which are a prerequisite for policy change (Kingdon 1984, 1995, 2002;
Meijerink 2005; Zahariadis 1992). In fact, in MS ‘focusing events and crises are
essential to the coupling process that facilitates the opening of policy windows’ (Wood
2006, p. 420). Hence it seems that in MS there is an ‘excessive role played […] by
external factors in determining the opportunities for change’ (Capano 2009, p. 22).
However, MS also acknowledges that a focusing event may as well be the tipping point
of a longer process and that ‘focusing events are not always so straightforward’
(Kingdon 1995, p. 96). Still, MS does not provide a conceptual approach to analyse
these processes and relies to a large extent on focusing events in the classical sense.
23
Theories of policy change
First-generation PET also relies heavily on external events as the key driver of policy
change (Baumgartner and Jones 1991, 1993; Boushey 2012; Breunig and Koski 2006;
Givel 2006, 2010; John 2003). External events may challenge policy monopolies, which
previously provided the source for stability, in the sense of highlighting attention,
mobilising interest groups, changing policy images (see above), and altering policy
venues (Baumgartner 2006; Baumgartner and Jones 1991, 1993, 2009; Givel 2006; True
et al. 2007a; Wood 2006). In this context, PET theorists refer to ‘focusing events’ and
‘positive feedback’ as important mechanisms of sudden and dramatic policy change
(Baumgartner and Jones 2002b). External events therefore play an important role in
PET and ‘for punctuated equilibrium theories of agenda setting, the role of shocks or
crises cannot be understated’ (Pump 2011, p. 2). However, second generation PET
offers additional explanations and is much less dependent on external events.
Particularly the introduction of the concept of friction provides a theoretical explanation
of more gradual processes during which pressures build up until they reach a tipping
point (see section 2.3.1).
As shown above, all three mainstream theories of policy change rely on external events,
although not exclusively, as a cause of major policy change. The problem of such an
approach is that it puts the focus on the key external events that create pressures for
change, but not on the complex search process that follows whereby actors actually
determine what changes to make (Campbell 2004). This is not to say that external
pressures including crises do not trigger policy change - they do. But ‘policy change
itself is still largely influenced by factors normally thought to be endogenous to the
subsystem’ (Williams 2009, p. 48). Also, there is evidence that despite a crisis and the
expectation of major policy change sometimes no or only minor change takes place
(Birkland 1997; Givel 2008; Nohrstedt 2008; Walgrave and Varone 2008). Scholars of
24
Theories of policy change
all the three theories of policy change discussed recognise the need for a better account
of the role of external pressures for policy change. Reviewing the ACF, Weible et al.
(2009, p. 128) admit that ‘there is much to learn about the intervening steps between an
external perturbation and major policy change’. Similarly, John (2003, p. 489) asks PET
scholars ‘what is the relationship between the nature of the policy input and the
character of the policy output’? On a similar note, Weir (1992) stresses that MS is
ahistorical because it does not sufficiently acknowledge the policy legacy of the past
and how that impacts on policy making in the present. Responding to such criticism,
Zahariadis (1999) acknowledges that more empirical research is required to better
understand how processes such as incrementalism and path dependency affect MS. This
is not to say that external events do not matter – there is plenty of evidence that they do,
and analysis became much more sophisticated in understanding the causal mechanisms
that relate external events such as crises to policy change (Birkland 1997; Hansen 2007;
Nohrstedt 2008; Walgrave and Varone 2008).
In contrast to ACF, MS, and PET, incrementalism does not depend on external events as
a trigger for major policy change. While incrementalism had been accused of being
excessively conservative (1992), Lindblom argued that incrementalism is well placed to
explain more substantial policy change as well because ‘a fast-moving sequence of
small changes can more speedily accomplish a drastic alteration of the status quo than
can an only infrequent major policy change’ (Lindblom 1979, p. 520). Also Weiss and
Woodhouse (1992, p. 261) maintain that a ‘series of small steps obviously can lead to
significant change if there are enough of them coming quickly enough’. With regard to
the theoretical pattern of policy change this is certainly the case, but incrementalism
does not offer a comprehensive causal explanation under which circumstances major
policy change in frequent small steps is likely to happen and when it is not. Rather,
25
Theories of policy change
incrementalism stresses that there is no reason why it should not happen while focusing
on mechanisms that prevent more abrupt and significant policy change.
In sum, ACF, (first-generation) PET, and MS focus on short-term exogenous events
when explaining policy change – how long-term pressures affect policy outcomes
remains underspecified. Incrementalism does not depend on external events, but fails to
provide a convincing account of mechanisms that can lead to major policy change. In
search for promising theories of long-term policy change the chapter now turns to two
theoretical approaches that provide the basis for the conceptual framework of this thesis.
2.3 A new lens on policy change
Looking for a theoretical approach that captures more than the impact of external events
or incremental changes, this thesis looks through a relatively new lens at policy change
by employing the concept of friction that was developed recently by PET scholars in
response to some of the criticism their work received (Baumgartner et al. 2009; Jones
and Baumgartner 2005, 2012). Friction can be used to understand the build-up of
gradual pressures resulting in significant policy change. Similar ideas have been
explored by scholars of historical institutionalism such as Paul Pierson (2004b). In
addition to friction, the literature on policy feedback (Béland 2010; Pierson 1993a;
Weaver 2010) is used and merged with the friction framework. In the following, the
concept of friction is introduced and supplemented with the policy feedback literature.
Based on the two literatures, a set of research questions and hypotheses are developed.
2.3.1 Friction
‘When policies change, they will shift in a disjoint and episodic manner; as a consequence,
policymaking will appear to be in a period of exception to the general rule of stability—or
simply responding to unspecified “exogenous forces.” But in fact the disjoint policy
26
Theories of policy change
responses are part and parcel of the same policymaking process that generated the periods
of stability.’ (Jones and Baumgartner 2012, p. 7)
This quote from Jones and Baumgartner shows that PET has developed and attempts to
provide a much more comprehensive account of policy change over time. Part of this is
the development of the concept of friction, which addresses the question of how longterm pressures affect policy outcomes.
The concept of friction
The concept of friction is based on the idea that policy makers (because of their
bounded rationality and cognitive limits) can only deal with a limited number of
questions at a time and the multiple institutional venues constrain and slow down the
policy response to policy inputs. Hence, there is a misbalance between policy inputs and
policy outputs. As a result, some issues are not dealt with and pressure starts to build up
in the system (Baumgartner et al. 2009; Jones and Baumgartner 2005). Baumgartner et
al. compare the mechanism to earthquakes:
Violent earthquake results from the friction and the associated buildup of pressure, not any
momentary increase on the forces pushing to overcome the friction. At any given time, the
response to the pressure is out of sync with the level of pressure applied: friction causes the
linkage between inputs and outputs of the system to be disproportionate - underresponse
because of friction, then overresponse in response to built-up pressures. (Baumgartner et al.
2009, p. 607)
Another metaphor for friction is a sandpile. The steady drop of sand grains on a flat
plate does not lead to a constant flow of grains falling off the plate. In contrast, usually
when one grain is added to the pile nothing happens and a sandpile builds up. But
sometimes when one grain is added to the pile the system collapses in rapid landslides.
27
Theories of policy change
This pattern is due to the friction of the sand and would not occur if material with no
friction was used (Jones and Baumgartner 2005).
Friction and the policy process
The policy process follows a similar logic of such ‘stick-slip dynamics’, Jones and
Baumgartner (2012, p. 8) argue. Policy makers need to prioritise some policy inputs
and neglect others. As a result, decision makers underrespond to issues if they remain
below a certain threshold and only concentrate on those areas where concern is great
requiring immediate attention. After an issue passes the threshold, over-response may
happen as a result of past negligence and the built-up pressures, a form of positive
feedback (Baumgartner and Jones 2002b). Therefore friction ‘is not an absolute barrier
to action, but rather a major hindrance. As it operates, pressure can mount, making
change when it does occur, more profound’ (Jones and Baumgartner 2005, p. 88). In
essence, friction is driven by negative feedbacks keeping the status quo, in other words
a homeostatic process preventing change (Baumgartner and Jones 2002b).
Jones and Baumgartner (2005) provide two causal explanations of why there is
disproportionality between policy inputs and outputs: cognitive limits of organisations
and institutional procedures.
Cognitive limits of organisations result in disproportional information-processing. There
are three components of Jones’ and Baumgartner’s (2005) disproportional informationprocessing model: First, the issues themselves to which one pays attention or ignores.
Second, the way these issues and their dimensions are interpreted. Third, the choice of
competing potential solutions one has to make. The concept is based on Simon’s (1983)
behavioural model of individual decision making and extended to organisations such as
governmental bodies. Because organisations can only process a limited number of
28
Theories of policy change
issues, most issues do not get the attention of major policy makers for most of the time.
Therefore, the ‘cognitive architecture of institutions creates an inevitable bottleneck of
attention’ (Walgrave and Vliegenthart 2010, p. 1149).
'Resistance built into the institutional structure of policymaking [my emphasis]’ (Jones
and Baumgartner 2012, p. 4) is a second reason for friction. Past legacies of institutions
can add friction due to sunk costs, long-term budgetary commitments, and bureaucratic
inertia. Another source of friction is that policies are often not terminated even if the
severity of a problem has declined significantly. Therefore, ‘rather than well-oiled,
smoothly churning cogs, political institutions are more like rusted gears - that is, they do
not move easily’ (Workman et al. 2009, p. 80). This makes the reallocation of resources
to other areas more difficult and the response is slow (Baumgartner et al. 2009).
The effect of friction can be compared to a bottleneck that limits the amount of policy
inputs to policy outputs. A good metaphor for the effect of bottlenecks can be found in
the transport literature (e.g. Vickrey 1969): Imagine a stretch of motorway between two
interchanges during the morning rush hour commute. On a relatively short route
segment road works take place, two-way traffic flow is still possible, but slowed down
from say 70 km/h to just 50 km/h allowing for a fixed maximum number of cars to pass.
This is the bottleneck - as long as the traffic flow remains below the capacity of the road
segment with the road works, there is no congestion. But once the traffic flow exceeds
the capacity of the bottleneck, traffic is slowed down, and when the number of cars
continuously surpasses the maximum capacity queues accumulate. In case of policy
change, the narrowness of the bottleneck (the degree of friction) determines whether
policy inputs result in policy outputs.
As a result of this, policy change is expected to follow patterns of PE:
29
Theories of policy change
If and when these issues have won the attention of the primary policymaking institutions,
errors have often accumulated, and punctuations must occur to “catch up” with a changing
reality. While institutions can parallel process numerous policy matters simultaneously, the
processing system is imperfect, and these imperfections lead to punctuations. (Jones and
Baumgartner 2005, p. 6)
Jones and Baumgartner (2005) refer to the importance of threshold effects i.e. situations
when the policy inputs (or signals as they call it) cross a limit and receive attention by
policy makers. Thresholds can either be reached by accumulation of pressures or very
strong external pressures (Jones and Baumgartner 2012).
However, Jones and Baumgartner (2005) stress that thresholds are contingent for two
reasons: First, new issues require agenda space and depending on how crowded the
agenda already is they only find that space in some situations. Second, different
institutions respond to policy inputs differently because they are biased (e.g. different
political parties). Hence for some institutions a threshold might be reached whereas
others only respond to a much higher threshold. Also, the capability of institutions to
respond to signals and generate policy change differs. While in some cases existing
institutions might be well equipped to deal with the issues arising, in other cases an
institution simply is not able to respond to a problem and even poses a barrier to policy
change.
Research gaps
There is a particular need for comparative work applying the concept of friction to
different institutional settings. John (2006, p. 983) asks ‘why are political solutions and
outcomes different in countries that share many background problems, and what is the
impact of different political institutions on policy choices?’ The concept of friction
30
Theories of policy change
might help answer some of these questions. Assessing the role of institutions for policy
change in PET theory remains one of the promising avenues for more research
(Baumgartner et al. 2009; John and Jennings 2010; Schlager 1999).
So far the concept of friction has mainly been used in quantitative studies and there is a
lack of qualitative work shedding light on the causal mechanisms of friction requiring
‘more in-depth fieldwork-based studies of actual policy processes as these are worked
out on the ground’ (Jones and Baumgartner 2012, p. 13). This is why Baumgartner et al.
(2009, p. 615) call for studies ‘using more qualitative methods to understand the
processes which lead to friction in greater detail, and why it is lower in some
institutions than in others’. In light of this, a comparative study seems to be particularly
promising, an approach this project incorporates.
Because the concept of friction is based on both positive and negative feedback
processes, another avenue for discovering causal mechanisms that lead to friction is the
wider literature on policy feedback. The thesis will explore potential feedback
mechanisms and merge those with the existing friction approach presented above in a
comprehensive conceptual framework.
2.3.2 Policy feedback
Obviously, the choice and calibration of policy instruments depends on politics
(Freeman 1985), but ‘new policies [also] produce new politics’ (Schattschneider 1935,
p. 288), a claim famously rephrased as ‘policies determine politics’ by Lowi (1972, p.
299). This is because ‘policy choices have political consequences’ (Pierson 1993b, p.
597) and ‘policies, once enacted, restructure subsequent political processes’ (Skocpol
1992, p. 58) such as rearranging resources and opportunity structures. Hence ‘policies
[…] have their own lives, their own internal logic and goals: they not only passively
31
Theories of policy change
adapt to external inputs, but indeed they actively influence external factors’ (Capano
2009, p. 27). In a nutshell, ‘policy feedback simply refers to how policies affect politics
over time’ (Béland 2010, p. 569). However, politics also determines the choice of policy
instruments and therefore the ‘direction of causation between policy and politics is […]
two-way’ (John 1998, p. 8).
The literature on policy feedback is vast and covers many areas. More recently, Béland
(2010) undertook a comprehensive literature review, 17 years after Paul Pierson (1993a)
wrote his seminal paper on policy feedback. According to Béland, there are at least
three types of policy feedback mechanisms: state building, interest group mobilisation,
and lock-in effects.
State building refers to an expansion of the administrative capacity of the state as a
result of implementing a new policy which requires additional skills and systems. Once
the administrative capacity of the state expands, future policies that require similar
capabilities are potentially more likely to be implemented compared to a situation where
no such capabilities exist (Pierson 1993a; Skocpol 1992). For example, the UK
implemented a domestic emissions trading scheme long before the start of the EU
Emissions Trading Scheme whereas Germany did not. As a result, the administrative
capabilities of how to run such a scheme already existed in the UK before the EU
Emissions Trading Scheme started in 2005. In Germany, in comparison, the government
departments and agencies had no such experience. As a result, the UK government
strongly supported the EU European Emissions Trading Scheme in contrast to Germany
where parts of the government and administration tried to prevent the start of the EU
European Emissions Trading Scheme until the very end (Engels et al. 2008). There were
of course other reasons for the different reaction of the two countries such as different
regulatory styles and traditions (Wurzel 2008).
32
Theories of policy change
Interest group feedbacks describe the effect that ‘policies provide both incentives and
resources that may facilitate or inhibit the formation or expansion of particular groups’
for example by creating ‘"spoils" that provide a strong motivation for beneficiaries to
mobilize in favo[u]r of programmatic maintenance or expansion’ (Pierson 1993a, p.
599). In other words, if the government establishes a new policy that benefits a certain
group, it is likely that the beneficiaries of that policy will not only push for keeping that
policy in place, but demand to expand the policy in order to increase the benefits they
enjoy. A good example is the case of the wind energy industry in Germany, which
benefited from the Electricity Feed-In Law introduced in 1991. The policy instrument
guaranteed wind power producers a fixed price per unit of electricity produced and
obliged utilities to accept power produced from renewable sources including wind.
However, in 1997 the Electricity Feed-In Law almost came to a close, but strong
pressure from the wind energy industry lobby mobilised public opposition resulting in
the maintenance of this policy instrument (Michaelowa 2005). Additional to providing
resources to interest groups, policies can also help the organisation and formation of
interest groups (Pierson 1993a) because ‘[p]rogram resources also create material
incentives for groups to band together in political activity’ (Mettler and Soss 2004, p.
62). Again, the wind power industry example can be used to illustrate such effects:
Because wind power in Germany was mainly deployed in structurally weak regions
where it created jobs and income, the wind energy industry successfully formed
alliances with local politicians, farmers, and trade unions, all which are very influential
in the political processes of increasing the share of renewable energy. This alliance
helped the wind energy industry a great deal to push government for the maintenance
and expansion of the Electricity Feed-In Law (later the Feed-In Tariff) (Michaelowa
2005). At the same time, other interest groups that do not benefit from a policy
33
Theories of policy change
instrument but are affected by negative consequences are likely to mobilise against the
introduction or expansion of such a programme. In fact, the utilities did mobilise against
the German policy on renewable energy but did not succeed (Lauber and Mez 2004).
Existing policies often cause lock-in effects that constrain, or even prevent, policy
change (North 1990, 1998, 2005; Pierson 1993a, 1994, 2000a, 2001, 2004b, 2009). This
is because ‘public policies operating in a context of complex social interdependence will
often generate increasing returns as well as high fixed costs, learning effects,
coordination effects, and adaptive expectations’ (Pierson 1993a, p. 608). In the energy
literature, lock-in effects are often restricted to technology lock-in and analysis focuses
on how technological and socio-economic aspects are intertwined creating powerful
systemic barriers to change (Unruh 2000). However, the same applies to policies: Often
new government agencies are created as a result of new political instruments, staff need
to be retrained, complex administrative procedures need to be established etc. Diverting
from the status quo implies potentially costly changes within government organisations.
Also, the regulated entities get used to a certain type of policy and learn how to deal
with it effectively. These factors make it increasingly difficult to replace or significantly
amend an existing policy instrument.
In addition to the three types of policy feedback described above, Béland suggests three
supplementary types of policy feedback mechanisms based on more recent literature:
private institutions, political participation, and ideational and symbolic legacies.
Ideational and symbolic legacies embedded in existing policies may influence
subsequent policies in multiple ways, for example by enabling or preventing actors to
use symbols, categories, and ideas that are represented in the policies that prevail. A
good example is the UK’s Renewables Obligation, a policy instrument which obliged
the energy utilities since 2002 to generate a certain proportion of energy from renewable
34
Theories of policy change
sources. For each MWh of renewable energy the utilities received a Renewable
Obligation Certificate that they could sell to other energy suppliers or brokers. The
Renewables Obligation was heavily criticised for not being effective and too costly
compared to alternative policy instruments such as Feed-In Tariffs (Mitchell 2009;
Mitchell et al. 2006; Toke 2005; Wood and Dow 2011; Woodman and Mitchell 2011).
Despite mounting criticism, the responsible Business Secretary, John Hutton, argued
that the RO ‘has overseen a rapid increase in renewable energy in the UK and we should
stick with what works’. One reason why the UK government decided to keep the RO
constituted the firm belief in market based instruments and the Renewable Obligation’s
fit with the dominant regulatory approach (Toke and Lauber 2007). Quoting John
Hutton again, this also comes out very clearly: ‘The Renewables Obligation is a genuine
market mechanism and that is why it has been successful’ (Hansard 2008, Column 527).
Hence, in this case lock-in occurred due to the firm rooting of the RO in marked-based
regulation obstructing the replacement of the policy instrument.
Private institutions providing resources to particular groups can also generate feedback
effects with relevance for public policy. A prominent study on private social
programmes and feedback was conducted by Hacker (2002) who showed how private
social schemes impacted on subsequent public policies. Political participation can be
stimulated by existing policies and create pressures for policy change. This is different
to top-down organised interest group mobilisation and refers to bottom-up political
participation of individuals (Campbell 2002a).
In addition to the types of policy feedback outlined by Béland, there are also negative
feedback mechanisms, i.e. processes as a result of policies that undermine the status quo
(Weaver 2010). The English novelist Arnold Bennett once said that ‘any change, even a
change for the better, is always accompanied by drawbacks and discomforts’ (Bennett
35
Theories of policy change
1912, p. 103). Drawbacks and discomforts of policies are precisely how negative
feedback mechanisms are understood. Opposition from organised vested interests
against the new policy regime is an obvious type of negative feedback (Schrad 2010). A
well-known example is the lobbying activities of the oil industry in the US with the aim
to undermine climate change policies.
Other negative feedback mechanisms are more subtle. For example, the German Feed-in
Tariffs for renewable energy are financed via electricity bills and due to the rapid
increase in deployment of renewable energy generation consumers pay an increasing
proportion of their bills for the support of Feed-in Tariffs (Frondel et al. 2010). While it
is too early to predict the political consequences, complaints about rising prices and the
negative impact on social equity has already been a focal point of recent debates in
Germany. At some point resistance against rising electricity prices might pose a serious
challenge to the Feed-in Tariffs model.
2.4 Combining punctuated equilibrium theory and policy feedback
literature
The challenge of applying both the friction model and the policy feedback literature lies
in the complexity and conditionality of the predicted policy processes. Hence ‘the real
analytical problem from the point of view of the framework’s theoretical coherence, is
the feedback interaction between the endogenous and exogenous sets of factors. The
feedback sequence and loops make the difference here’ (Capano 2009, p. 17).
This section starts with assessing the compatibility of the friction approach and the
policy feedback literature before it develops the conceptual framework of this thesis.
2.4.1 Compatibility of friction and policy feedback literature
When combining different theories in a common framework as done in this thesis, the
compatibility of the theories needs to be considered. Although they have different
36
Theories of policy change
origins, their basic assumptions are not that different. With regard to policy feedback,
however, the definitions used are inconsistent and require clarification.
Origins
Policy feedback mechanisms are mainly rooted in the theory of historical
institutionalism (Campbell 2004; Pierson 1993a, 1994, 2004b; Pierson and Skocpol
2000; Skocpol 1979, 1992; Steinmo et al. 1992; Thelen and Steinmo 1992; Weir 1992),
whereas the friction concept originates from the literature on policy change and
particularly PET (Baumgartner et al. 2009; Baumgartner et al. 2006; Baumgartner and
Jones 1991, 1993, 2002a, 2002b, 2009; Jones and Baumgartner 2005, 2012; True et al.
2007a; Walgrave and Varone 2008; Walgrave and Vliegenthart 2010).
In the past, the policy change literature (including the three main theories ACF, MS, and
PET) and the historical institutionalism literature have been quite separate, the main
reason probably being that policy change scholars looked at policies and historical
institutionalists looked at institutions, although this was not always that clear-cut.
However, the fact that the two literatures have been detached does not mean they cannot
be used collectively and scholars of both theories indicate complementarity. For
example, Pierson makes a convincing case for treating ‘public policies as institutions’,
encouraging historical institutionalists to look at policies as well (Pierson 2006). Also,
historical institutionalists such as Hall (1993, p. 277) write about ‘‘punctuated
equilibrium’ that often applies more generally to political change’. Reviewing Hall’s
seminal 1993 paper on policy change, Baumgartner stresses that PET theorists ‘reach
the same conclusions from different approaches, and to demonstrate the validity of
similar ideas with very different methodologies’ (Baumgartner 2011, p. 5).
37
Theories of policy change
Basic assumptions
With regard to the basic processes of the friction concept, PET refers mainly to two
phenomena: First, to the bounded rationality of individuals and organisations and their
limited cognitive ability to process information, and second, to institutional friction due
to the limited capability of institutions to deal with issues. Both mechanisms lead to
failure to translate policy inputs into policy outputs (Baumgartner et al. 2009; Jones and
Baumgartner 2005, 2012). Historical institutionalists
also stress the inability of
individuals to make purely rational decisions taking into account all possible effects of
their actions (Hall 1983; Katznelson 1998; Steinmo 1989; Steinmo et al. 1992). In fact,
historical institutionalism is heavily influenced by the work of organisational theorists
(Koelble 1995), particularly the work of March and Olsen (1984, 1989) who draw on
the work of Simon (1957, 1959) on bounded rationality. Similarly to the friction model
of Jones and Baumgartner, Pierson (2004b) stresses the cognitive limits of actors and
the multiple biases affecting decision-making. Furthermore, he argues that individuals
often (although not always) have short time horizons not considering long-term
consequences. However, historical institutionalists do attribute explanatory power not
only to cognitive limits, but also to the institutions setting the context in which decisions
are made (Katznelson 1998). To some extent, friction does capture the effect of
institutions with regard to their capability to deal with issues, but the emphasis of
friction lies on cognitive limits whereas historical institutionalists stress institutional
aspects.
Furthermore, historical institutionalism spent a great deal of effort on explaining why
institutions and policies persist over long periods of time, whereas PET focuses on
abrupt change while acknowledging that institutions and policies may persist for very
long periods of time (Jensen 2009).
38
Theories of policy change
So although the two approaches do have quite different origins and emphasise particular
elements, there is no apparent fundamental incompatibility in terms of the basic
assumptions.
Definitions of feedback mechanisms
While both literatures refer to positive and negative feedback mechanisms, they use the
terms in opposite ways:
Positive feedback is referred to by the historical institutionalism literature as processes
that stabilise and reinforce existing policy regimes, which may eventually result in lockin situations (Campbell 2002a; Mettler and Soss 2004; Pierson 1993b, 2004a; Skocpol
1992). PET turns this definition on its head and describes positive feedback as a
mechanism that drives rapid policy change rather than reinforcing existing policy
regimes. Hence positive feedback ‘can lead to a cascade or a spiral of subsequent events
that dramatically change the status quo’ (Baumgartner and Jones 2002b, p. 15).
Historical institutionalists define negative feedback as the ‘consequences of policy that
tend to undermine rather than reinforce the political, fiscal or social sustainability of a
particular set of policies’ (Weaver 2010, p. 137). In contrast, PET scholars use the term
negative feedback for processes that are self-correcting and status quo stabilising.
According to PET, the ‘key element of any negative feedback system is simply that the
system reacts to counter-balance, rather than reinforce, any changes coming in from the
environment’ (Baumgartner and Jones 2002b, p. 9).
Hence the two literatures use opposite terms for processes that lead to an expansion and
contraction of a policy regime respectively (Table 1).
39
Theories of policy change
Table 1: Positive and negative feedback in Punctuated Equilibrium Theory and historical
institutionalism
Theoretical approach
Punctuated Equilibrium
Historical institutionalism
Positive
undermining of policy regime
expansion of policy regime
Negative
expansion of policy regime
undermining of policy regime
Type of policy feedback
In order to conceptualise policy feedback in a consistent manner I stick to the natural
science definition of positive and negative feedback: Positive feedback exacerbates
particular changes whereas negative feedback dampens and inhibits change (Holling
1973).
However, it is important to think of positive and negative feedback mechanisms not as
one-directional pressures, but as forces that can both exhilarate and retard policy
change. While positive feedback effects (negative in PET) are often thought of as
change inhibiting factors, they can also put further pressure on policy makers to do
more of what they do (for example because it benefits a certain interest group
demanding an expansion of current policies). Similarly, negative feedbacks (positive in
PET) do not necessarily lead to policy change: If, for example, a new policy instrument
is introduced and negative feedback effects are apparent immediately, these effects may
simply result in reverting back to the status quo (Patashnik 2008).
It is important not to think of change or no change as a generally positive or negative
development. In some instances policy stability (no change) is important to allow for a
policy being effective whereas in other situations change is a prerequisite in order to
increase the efficacy of a policy.
40
Theories of policy change
2.4.2 Conceptual framework and hypotheses
The conceptual framework of this project is based on the literatures of PET and
historical institutionalism. Coming back to the introduction of this chapter, the
conceptual framework is supposed to illuminate how long-term pressures (both external
but also internal to the policy subsystem) impact on policy change. The following
elements are integrated in the framework:

Pressures external to the subsystem are conceptualised as drivers from a wide
range of possible sources that put pressure on the policy subsystem to change.
These pressures may only operate for a very short period of time, for example a
few days, but may also last several years or even decades. Short-term events are
referred to as systemic perturbations whereas long-term pressures are called
subsystem spillovers (Williams 2009).

Policy inputs can be defined as ‘information relating to the state of affairs “out
there” that actually comes to the attention of government’ (Baumgartner et al.
2009, p. 605). Information comes in various shapes such as indicators (e.g. new
scientific research on environmental problems), focusing events (e.g. an energy
crisis), and feedback from existing policies (Kingdon 1984, 1995). It can be
provided by the media, lobby groups (including non-governmental organisations
(NGOs)), experts, and government agencies, amongst others (Baumgartner et al.
2009).

Policy outputs include budgets, new regulations, modification of existing
regulations, and other authoritative decisions (Baumgartner et al. 2009). Hence
they comprise any changes in policy goals, objectives, or instruments (Hall
1993). The primary focus of this thesis lies on two selected policy instruments,
namely the SO and the CBRP, and changes in their settings over time.
41
Theories of policy change

Institutional capability puts the process of friction at the core of the conceptual
framework – the ability of the system to respond to pressures and policy
feedback is a function of the degree of friction. The larger the degree of friction,
the higher the proportion of policy inputs which are not translated into policy
outputs.

Policy feedback encapsulates both expansion driving and status quo
undermining effects from existing policies (in the case of this thesis the selected
policy instruments). In the conceptual framework these processes are treated as
policy inputs.
Figure 2 schematically represents the conceptual framework and can be thought of as a
flow diagram. The size of the arrows is indicative only and shows the magnitude of
pressures, inputs, and outputs. Depending on the context, the importance of the different
processes will be different.
Figure 2: Conceptual framework
42
Theories of policy change
2.4.3 Hypotheses
Based on the principles of regression analysis some scholars argue that ideally, a set of
clearly defined hypotheses predicting the causes, and causal mechanisms of change
should be developed and then be tested (King et al. 1994). Such an approach is based on
the ‘linear-additive view of causality, which assumes that there is a clear separation
between independent and dependent variables’ (Capano 2009, p. 17). Outspoken
advocates of this view called for more formal models and even demanded to ‘put an end
to […] hopelessly vague theories about the policy process’ (Dowding 2001, p. 102).
The underlying assumption is that if you cannot formally test a theory, it is obsolete.
However, ‘there is more than one way of doing political science’ (Marsh and Smith
2001, p. 528) and different approaches make different epistemological and ontological
assumptions. More recent theories of policy change have become increasingly complex;
variables are interdependent and do not operate in a straightforward linear manner
(Capano 2009; Hall 2003; John 2003; Marsh and Smith 2001; Pierson and Skocpol
2002; Rayner 2009). It is for this reason that Peter Hall (2003, p. 375) argued that ‘the
ontologies of comparative politics have substantially outrun its methodologies’. This is
not to say that approaches based on regression analysis are not valid, but that depending
on the context of the investigation different types of methodologies are more or less
suitable.
Because the variables looked at in this project are in fact interdependent and related to
each other in multifaceted ways, a formal model categorising the causal factors into
independent, dependent, and intermediary variables is not appropriate. In such cases a
different approach called systematic process analysis (or process tracing) 2 , is
2
See the detailed description of this method in the methodology chapter.
43
Theories of policy change
recommended (Hall 2003, 2008). The steps of such an approach are to formulate a set of
theories and make predictions of causes and the processes by which these causes have
an effect on outcomes. Hall suggests using more than one theory and deriving rival
hypotheses in order to assess a variety of potential explanations and not to narrow the
scope of the investigation from the outset. Such a strategy addresses the danger that if
using one theory one just identifies the facts that fit with that theory. Following Hall’s
approach, a number of hypotheses are developed in the next section. Each hypothesis is
contrasted with a counterhypothesis.
Pattern of policy change in response to gradual pressures
There are various ways in which policy might change in response to gradual pressures.
Recent PET concepts such as friction assume that the system responds in a non-linear
fashion to such forces of discontent, when over long periods of time, policy inputs (i.e.
demands for change) are not in line with policy outputs (changes in policies). Pressures
build up over time as a result of inadequate responsiveness in the system due to
institutional friction. At some point a tipping point may be reached, and rapid policy
change may be triggered. In such situations, the system is expected to over-respond.
Following the above, the first hypothesis is as following:
H1/1: Long-term gradual pressures lead to changes to policy instruments that
follow a pattern of punctuated equilibrium. The scale of change is substantial
compared to the status quo.
The rival hypothesis to this is that policies change gradually in line with the pressures,
keeping a balance between the policy inputs and outputs:
H1/2: Policy instruments change incrementally as a result of gradual pressures.
The scale of change is minor compared to the status quo.
44
Theories of policy change
Friction
PET predicts that friction leads to the building up of pressures over time as a result of
bottleneck effects. The hypothesis thus is:
H2/1: Friction significantly limits the capability of institutions to respond to
gradual pressures and creates a misbalance between policy inputs and policy
outputs. The result is the accumulation of pressures.
While the friction concept is compelling, there may as well be no evidence for friction
of any kind and policy inputs are linearly translated into policy outputs.
H2/2: Friction does not limit the capability of institutions to respond to gradual
pressures and there is a balance between policy inputs and policy outputs.
Pressures do not accumulate as a result.
PET also provides two causal mechanisms that lead to friction:
H3/1: Friction is caused by a) the limited ability of institutions to deal with
information i.e. policy inputs, and b) prohibitive institutional structures that do
not allow for effective translation of policy inputs into policy outputs.
An alternative hypothesis is:
H3/2: Friction is caused by other factors.
Feedback
Both PET and historical institutionalism emphasise the importance of feedback
mechanisms, including positive and negative policy feedback.
H4/1: Existing policy instruments create positive and negative feedback
processes. Positive policy feedback leads to an expansion of the policy whereas
45
Theories of policy change
negative policy feedback results in a dampening of the expansion or contraction
of the policy.
However, one could also argue that these processes are absent and do not contribute to
change of policy instruments. Rather, politics independent of the policy instrument may
determine its modification.
H4/2: Existing policy instruments do not create positive and/or negative
feedback processes. Policy change is a result of political processes not
influenced by existing policies.
Conclusions
This chapter started with a dilemma: An initial screening of the cases under
investigation showed that the drivers of policy change appear to be long-term pressures
(e.g. the perceived impact of climate change) or feedback effects of the policy itself
(e.g. increasing reliance on public expenditure). However, the main theories of policy
change either seem to suggest that external events such as crises account for substantial
policy change or, in the case of incrementalism, that significant policy change is
unlikely to take place. In contrast, this chapter proceeded with a search for theories that
could help explain the impact of more long-term pressures on policy change. It started
by critically discussing four theories of policy change, namely incrementalism, ACF,
MS, and (first-generation) PET, and demonstrated that the theories that explain major
policy change rely heavily on external crisis type events. In search for more
comprehensive theoretical accounts two types of literature were identified: Recent
developments in (second generation) PET resulted in the notion of friction which
conceptualises the impact of long-term policy drivers and the building up of pressure
over time eventually leading to policy change. This literature was blended with theories
46
Theories of policy change
of policy feedback mechanisms coming out of historical institutionalism and a critical
assessment of their compatibility showed that in general the two types of literature are
complementary. Subsequently, the conceptual framework of this thesis was developed
based on both the literature of friction and policy feedback. Finally, the chapter derived
a number of rival hypotheses in order to guide the research and focus the analysis. The
logical next step in the following chapter is to establish the methodology used in order
to evaluate the hypotheses and to operationalise the different elements of the conceptual
framework.
47
Methodology
3 Methodology and research design
The investigation of the truth is in one way hard, in another easy. An indication of this is found
in the fact that no one is able to attain the truth adequately, while, on the other hand, we do not
collectively fail, but everyone says something true about the nature of things, and while
individually we contribute little or nothing to the truth, by the union of all a considerable
amount is amassed.
Aristotle (Metaphysics 993). Quoted in Feyerabend (1981)
‘There is no such thing as a social science’ some argue (Hutchinson et al. 2008), believing
in the premise that scientific methods and approaches are inappropriate for the study of
social phenomena, although a significant amount of social science is based on precisely
these methods. Even if one accepts the notion that there is no such thing as social science,
the question remains, however, which sort of method is appropriate for studying politics.
Being trained as a scientist, I was used to fairly straightforward methods such as statistics
and relatively few controversies around the general approach to scientific analysis.
Looking at the social sciences, the breadth of viewpoints on the question of methodology
and the divide between different schools of thought is striking. Given the range of possible
choices, it is the task of this chapter not only to present, but also to position and justify the
methodology selected for this study.
Considering that many of the disputes about method stem from disagreements about
ontology and epistemology, this chapter starts by locating this thesis in ontological and
epistemological terms. Turning to more practical matters, the chapter discusses the case
study approach used in this project and provides reasons why such a technique has been
chosen and not others. It will then present the method of systematic process analysis,
48
Methodology
which will be used to analyse the case studies. Subsequently, a justification for the
selection of countries and the time horizon looked at in the research is provided. This is
followed by an introduction to the methods used including semi-structured interviews and
document analysis. Finally, the chapter concludes with operationalising the key theoretical
concepts developed in the previous chapter.
3.1 Ontological and epistemological assumptions
In order to enable others to critically assess this thesis the main ontological and
epistemological assumptions need to be spelled out. Unfortunately, this part is often
missing in doctoral theses and the ontological and epistemological assumptions can be
second-guessed only from the theoretical frameworks employed and the methods used. By
being clear about ontology and epistemology, an approach recommended by scholars
working on policy change (Capano 2009), the suitability of the approach taken in this
thesis can be judged more adequately.
3.1.1 Ontology
Ontology ‘relates to the nature of the social and political world’ (Hay 2007, p. 117), it
‘refers to the claims or assumptions that a particular approach to social enquiry makes about
the nature of the social reality – claims about what exists, what it looks like, what units make it
up and how these units interact with one another’ (Blaikie 1993, p. 6).
By choosing a certain theoretical framework or by developing a specific research design,
implicit ontological assumptions about the political world are made such as structure and
agency, institutions, material and ideas, political actors, and in particular the causal
relationships in this world. Hays argues that
‘no political analysis has ever been ontologically neutral; rather fewer political scientists are
prepared to proceed on the basis of this once unacknowledged and unchallenged presumption’
(Hay 2006, p. 78).
49
Methodology
Therefore the ontological presumptions of the theoretical framework adopted in this thesis
are laid open. The two main streams of theory used in this thesis are recent concepts of
PET and the literature on policy feedback, which is based in historical institutionalism.
PET places the policy process ‘on a dual foundation of political institutions and boundedly
rational decisionmaking’ (True et al. 2007b, p. 156), both small and large-scale change
comes from interaction of multilevel political institutions and behavioural decision
making. Similarly, historical institutionalism ‘see[s] the world not as a terrain marked by
the operation of timeless causal regularities, but as a branching tree whose tips represent
the outcomes of events that unfold over time’ (Hall 2003, p. 385) opposed to equilibrium
concepts of rational choice theories. Variables are not independent but often bundled
depending on the particular context. The sequence of events matters and has a significant
impact on the outcomes of processes. Therefore, a historical institutionalist perspective
rejects functional explanations which claim that institutions only exist because they fulfil a
certain function, often institutions evolve in a complex causal chain over time (Pierson and
Skocpol 2000). In sum, this thesis looks at political processes from a perspective that
assigns particular importance to institutions, actors, time and sequence.
3.1.2 Epistemology
Ontology conceptualises our assumptions on the nature of the world; epistemology relates
to ‘to what we can know about it’ (Hay 2007, p. 117).
This thesis adopts a critical realist epistemological perspective. It follows Cruickshank
(2003, p. 2) who supports the view that ‘research [is] from the very start influenced by
assumptions’ and research is ‘about gaining knowledge of a reality that exists
independently of our representations of it’ (p. 3). That knowledge then is interpreted
through the theoretical lens of the researcher. Because of that limited access to knowledge
50
Methodology
of a reality, ‘causal mechanisms can exist independently of our knowledge of them’
(McAnulla 2005, p. 32).
The theoretical framework adopted is consistent with a critical realist perspective. As
Marsh convincingly shows, historical institutionalism can be located in a critical realist
position (Marsh 2008). This is probably less clear for PET and there are few if any
references to epistemology. However, PET assumes limited ability to process information
and cognitive limits of actors (Jones and Baumgartner 2005), which surely must also apply
to PET scholars placing PET comfortably within a critical realist perspective. Furthermore,
acknowledging the limitations of theoretical predictions, Jones and Baumgartner stress that
‘the most important aspect of a theory or framework is not whether it is right or wrong, but
the extent to which it is fruitful; that is, the extent to which it stimulated further research’
(Jones and Baumgartner 2012, p. 1). Again, this statement suggests a critical realist
perspective.
Both historical institutionalism and PET stress the importance of time when investigating
policy change (Pierson 2000b; True et al. 2007b). This thesis carries out an analysis in line
with the widely accepted presumption that studies looking at policy change should cover
several years or even decades (Sabatier 1993). Paying particular attention to the dimension
of time is an important feature of a critical realist approach and ‘[t]here cannot be any
[critical realist] analysis of change without reference, whether implicitly or explicitly, to a
conceptualization of time’ (Bates 2006, p. 145).
Furthermore, the importance of causal mechanisms is one of the key concerns of this thesis
trying to link empirically observable causes to effects by better understanding the way
these causes have an impact on outcomes. This is at the heart of critical realism (Bhaskar
1978). Finally, critical realism is also reflected by the method of investigation chosen for
51
Methodology
this project, it follows an approach called systematic process analysis, which is firmly
rooted in the critical realist perspective (Hall 2008).
3.2 Case study approach
This thesis aims to provide an advanced understanding of how long-term pressures affect
policy change putting the emphasis on the causal mechanisms associated with those
changes. Having to choose between qualitative or quantitative methods, or a mixture of the
two, advice given by Fischer et al. is taken into account:
‘the key to explaining how [policy] change comes about has to be grounded in a detailed
contextual examination of the circumstances at play in specific cases. For this purpose
quantitative methods have to take a back seat to qualitative research’ (Fischer et al. 2003, p.
108).
Also Bennett argues that case studies are particularly well suited for the
‘explanation of a sequence of events that produce a particular historical outcome in which key
steps in the sequence are in turn explained with reference to theories or causal mechanisms’
(Bennett 2004b, p. 21).
This is precisely what this research is about by asking questions focusing on the drivers
and causal mechanisms of policy change. Therefore, this thesis will use the case study
method in order to evaluate the hypotheses derived from the theory and answer the
research questions posed above.
The case study method has other advantages compared to a quantitative statistical
approach in a number of areas (Collier 1993; Mahoney 2007). Case study methods can
operationalise qualitative variables, identify heuristically new or omitted variables and
hypotheses, and examine potential causal explanations within particular cases (George and
Bennett 2005). This is not to say that quantitative methods are inferior, the case study
52
Methodology
method just has a different focus and is better suited to explain phenomena that involve
complex causal chains.
Using Lijphart’s (1971) well-known categorisation of different types of case studies, the
type of case study this thesis conducts can partly be classified as theory-confirming/
theory-infirming (because it tests hypotheses based on existing theories), and partly as
hypothesis-generating (because it adjusts existing and proposes new hypotheses).
3.2.1 Potential pitfalls of the case study method
Case studies run the risk of selection bias and there are tradeoffs concerning the number of
cases, hypotheses and variables (Bennett 2004b). The case study method has been
criticised for suffering from the degrees of freedom problem, which refers to a problem in
statistics when the number of independent variables is greater than the number of cases
(Goldstone 1997; King et al. 1994; Lijphart 1971). However, as Bennett (1991)
convincingly argues, the concepts of case, variable and observation cannot simply be
transferred from statistics to qualitative case study based research. Mahoney (1999; 2000a)
presents three strategies used in the case study method in order to reduce the problems of a
small number of cases: nominal comparison, ordinal comparison, and within-case analysis.
Nominal comparison refers to John Stuart Mill’s method of agreement and difference (Mill
1843). Based on Mill, the Most Similar Systems Design compares similar cases with
different political outcomes. Such an analysis can reveal those explanatory factors (e.g.
causes of policy change) that differ across similar cases (e.g. countries with similar level of
economic development, similar political systems etc.) and produce different outcomes (e.g.
different degrees of policy change). The Most Different Systems Design compares
countries that share few common features but show similar political outcomes (Landman
2008). However, Mill himself pointed towards the difficulties of using the method of
53
Methodology
agreement and difference in practice. Mill’s methods work well when several demanding
requirements are fulfilled: First, all causal relevant variables must be identified prior to the
empirical research. Second, cases representing all potential causal paths must be available
for study. Finally, the causal relation of the independent and dependent variable needs to
involve only one condition, and be either sufficient or necessary (George and Bennett
2005). These conditions are very hard to meet in practice. In particular, the multiple
causality of outcomes i.e. the equifinality poses a challenge to research design. Different
causal mechanisms can lead to similar outcomes making it very difficult to specify the
causes at work. Also, if not all causal paths are identified there is the danger of ‘spurious
correlation’ (Mahoney 2007). Spurious correlation refers to the phenomenon when an
independent variable seems to drive a dependent one but in fact both are dependent on an
antecedent variable that has been omitted.
Ordinal comparison is based on Mill’s method of concomitant variation and ranks the
cases based on the degree to which explanatory factors and expected outcomes are present.
Each case is scored in terms of the presence of the explanatory variables and the outcomes.
By exploring the covariation between causes and outcomes the degree of causality can be
assessed (Mahoney 2000a). Applied to the research on policy style change, covariation
between the causes of policy style change (explanatory variables) and change of policy
style (outcome) is tested. The higher the degree of covariation, the more plausible it seems
that the hypotheses prove an adequate explanation. However, deviation from the expected
pattern does not necessarily mean that a hypotheses can be rejected (Mahoney 1999).
Often ordinal comparison does not provide a clear basis for eliminating potential
explanatory variables because causal patterns are assumed to be partial and probabilistic
(Mahoney 2000b).
54
Methodology
A third strategy that can be used to prevent potential pitfalls of small-n case studies is the
within-case analysis. This approach deals with some of the problems of nominal and
ordinal comparison. It is different from nominal and ordinal comparison in terms of the
level of aggregation: both the ordinal and nominal approach entails aggregated
comparisons across cases whereas within-case analysis involves disaggregated
comparisons within cases (Mahoney 2000b). Due to the central research question of this
thesis, within-case analysis is deemed most suitable to generate the results needed.
3.2.2 Within-case analysis as a means to avoid pitfalls
Within-case analysis focuses not on the analysis of variables across cases but on the causal
mechanisms in a single case. Doing within-case analysis can prevent spuriousness which
can be a problem in both nominal and ordinal comparison. Within-case analysis can
compensate for the limitations of small-n studies by examining multiple features of a case.
By doing so, multiple observations of similar processes (e.g. processes of policy change)
are possible within one case leading to greater validity of the results (Mahoney 2000a). In
the context of a comparative study,
‘the results of individual case studies, each of which employs within-case analysis, can be
compared drawing them together within a common theoretical framework without having to
find two or more cases that are similar in every respect but one’ (George and Bennett 2005, p.
179).
Hence within-case analysis does not need to follow Mill’s method of agreement and
difference. In a later section of this thesis a discussion of the implications of choosing a
within-case analysis approach on case selection is provided.
There are various methods that can be chosen when doing within-case analysis and it is
important to be clear which method is used when carrying out the analysis. One of the
methods that is particularly attractive in the context of this thesis is systematic process
55
Methodology
analysis. Systematic process analysis is probably the most important tool when the aim is
to uncover causal mechanisms and reconstruct historical events or sequences of events
(George and Bennett 2005) which is the aim of this thesis. Therefore, this thesis will
employ systematic process analysis to analyse the cases chosen.
3.3 Systematic process analysis
Systematic process analysis, a term coined by Hall (2008), is one of the key tools available
when doing within-case analysis. Some refer to it as process tracing (Bennett 2004a;
George and Bennett 2005; George and McKeown 1985), others use the label pattern
matching (Campbell 1975). Systematic process analysis can be particularly useful to
employ theories
‘whose conceptions of the causal structures underlying outcomes are at odds with the
assumptions required for standard regression techniques and conventional comparative method
to provide valid causal inferences’ (Hall 2003, p. 375).
Systematic process analysis can be used when several theories refer to different causal
mechanisms to explain the same phenomenon (Hall 2008). Hence systematic process
analysis is not just assessing the causes of a phenomenon but also the causal process
whereby these causes operate (Bennett 2004a; George and Bennett 2005). Such a method
contains a deductive and an inductive element: It is deductive because it evaluates
hypotheses based on existing theories and inductive because the research might come
across unexpected explanations and variables overlooked so far (Bennett 2004a).
The following steps are part of systematic process analysis (Hall 2008):
1) Theory formation: approach the case with a range of theories and potential causes,
test one theory against another;
56
Methodology
2) Deriving predictions: develop predictions that are consistent with one theory but
inconsistent with its principal rivals;
3) Making observations: observation of correlation as in conventional comparative
method but also on process whereby an outcome is caused; and
4) Drawing conclusions: assessment to which extent the theories explain the outcome.
Steps 1 and 2 are already carried out in the theory chapter; It starts from a critical review
of the three most prominent theories on policy change and establishes a new lens on policy
change that takes account of more gradual long-term pressures. Based on existing
theoretical frameworks a number of hypotheses are derived. In addition, as suggested by
l ser and Laudel (2004), a number of research questions have been developed to guide
the research.
Step 3 will involve an indepth-analysis of the processes that led to policy change paying
particular attention to long-term pressures. Key long-term pressures will be identified as
well as short-term events that caused policy change, making sure more conventional
explanations of policy change are also covered in the analysis. When a factor is identified
as being an important causal driver the causal mechanisms by which this factor affects the
outcome are investigated further. A detailed overview of the methods used to make
observations is provided in section 3.5.
Step 4 will use a rigid method of analysis that is called Qualitative Content Analysis. This
will link the results of the qualitative research to the theoretical predictions made during
steps 1 and 2. Details of this approach are presented in section 3.5.4.
3.4 Selection of cases
Selecting the cases for the investigation is not straightforward. There is a lively debate on
the selection of cases for small-n studies. As outlined above, the case study method has
57
Methodology
been criticised for having a degrees of freedom problem, when the number of independent
variables is greater than the number of cases (Goldstone 1997; King et al. 1994; Lijphart
1971). However, as Bennett (1991) convincingly argues, the concepts of case, variable,
and observation cannot simply be transferred from statistics to qualitative case study based
research. First, within one case multiple observations of the same phenomenon can be
made. Second, the case study method allows an in-depth investigation of the causal
mechanisms of a few cases in contrast to merely testing whether the independent variable x
can explain the independent variable y statistically. This thesis employs the approach of
systematic process analysis, a method that ‘is fundamentally different from methods based
on covariance or comparisons across cases’ (George and Bennett 2005, p. 207). Systematic
process analysis does not rely on establishing causation through comparison and
covariation. Therefore, the critique of selection bias is misguided when it comes to studies
employing methods such as process analysis (Bennett and Elman 2006).
When trying to uncover causal mechanisms cases should not be chosen to be statistically
representative but because they display a high degree of the phenomenon that is
investigated (Pettigrew 1990). This thesis followed a strategic sampling approach when
selecting the cases.
Although a single case study can provide interesting insights regarding the causal
mechanisms, ‘increasing the number and diversity of the cases increases the investigator’s
confidence that the causal process observed is not idiosyncratic to one of them’ (Hall 2008,
p. 315). This is why a comparative approach is chosen for this thesis.
3.4.1 Geographical scope
This thesis investigates the principal home energy efficiency policies of Germany and the
UK. The reasons for choosing policies of these countries are manifold:
58
Methodology
First, both countries have been very active in the area of energy efficiency policy. They
have had a lot of change in the energy efficiency domain and there is sufficient material
that can be used to evaluate the hypotheses and predictions on policy change. It is expected
that the dynamics of energy efficiency policy in the two countries provide an excellent
case for applying theories of policy change.
Second, selection of cases with background knowledge allows a stronger research design
and cases can be selected with an expectation whether they will be valuable for theory
testing and/ or making (George and Bennett 2005). Both the author and the supervisors of
the thesis have preliminary knowledge of energy efficiency policy in both countries.
Third, by choosing Germany and the UK some degree of control for several critical
variables can be established. For example, domestic energy consumption per household is
similar in both countries. Also, the political system in both countries is democratic. Energy
markets in both countries are liberalised. Furthermore, both countries have strong climate
change targets with energy efficiency as a strong pillar.
Fourth, the two countries have quite distinct institutional structures. One of the research
questions is how institutional differences affect how long-term pressures impact on policy
change. This includes obvious formal differences such as Germany being a federal state
and the UK not, or that there are different agencies operating in the two countries.
3.4.2 Time horizon
Not just the geographical scope but also the time horizon needs to be carefully selected.
When selecting the cases for this study, three aspects with regard to timescales were
considered.
First, there is widespread agreement that policy dynamics should be studied over periods
of several years or even decades (Sabatier 1993).
59
Methodology
Second, the availability of data is likely to be inferior for periods further in the past than
for more recent periods. For example, older documents are not published online or might
not be available altogether. Also, it is more challenging to identify interviewees who were
observers of events that happened more than 20 years ago and even if they could be found
their memory of those events may be poor (Wengraf 2001).
Third, significant policy change occurred after the 1992 Rio Earth Summit and new
gradual pressures such as climate change really started to have an effect in the 1990s. The
1980s do not show similar turbulences in energy efficiency policy (see analysis below). As
outlined above, it is sensible to select cases that display a high degree of the phenomenon
under investigation (Pettigrew 1990). Therefore the time horizon is set after 1990.
3.4.3 Unit of analysis
When selecting the unit of analysis one needs to make sure ‘that the site and unit of
investigation are suitable for the type(s) of problem(s) that shall be investigated [emphasis
in original]’ (Diefenbach 2009, p. 879).
When approaching policy change, one needs to be clear which elements of policy are
actually looked at. Scholars focusing on policy change often took ‘policy’ as a dependent
variable of their research without specifying what they mean by policy (Heclo 1976; Rose
1976). Peter Hall (1986, 1993) made an attempt at decomposing the term ‘policy’ into its
different elements, namely the goals of policy making, the instruments deployed to attain
those goals, and the setting of those instruments. Hall argued that different types of change
occur at the various levels of policy in a different manner. Hall’s typology of policy
elements is widely used in the policy change literature.
However, although Hall significantly advanced the understanding of policy change and in
particular the conception of policy as a dependent variable, Hall’s model struggles to
60
Methodology
capture the difference between ends and means as well as the distinction between more
abstract goals/ logics and more specific settings/ calibrations (Cashore and Howlett 2007;
Howlett and Cashore 2009). There have been attempts to reconceptualise policy as a
dependent variable with the aim of providing a more coherent framework for analysis that
allows a better understanding of the elements affected by policy change, the direction of
change and the interaction among the different elements. Building on Hall, Howlett and
Cashore (2007; 2009) propose a six-dimensional taxonomy of policy components
disaggregating between ends and means as well as three different levels of abstraction
(from high levels of abstraction to specific on-the-ground measures). The following table
presents this framework and provides examples related to energy efficiency policy to
illustrate what is meant by each element:
Table 2: Taxonomy of policy components with examples
High level abstraction
Policy ends or aims
Goals: What general
types of ideas govern
policy development?
(e.g. energy security,
economic efficiency,
climate change
mitigation)
Policy means or tools
Instrument logic: What
general norms guide
implementation
preferences?
(e.g. preference for
market based
instruments or coercive
instruments, preference
for cost-effective
instruments)
Programme level
operationalisation
Objectives: What does
policy formally aim to
address?
Specific on-the-ground
measures
Settings: What are the
specific on-the-ground
requirements?
(e.g. reducing total
energy consumption by
x%, reducing carbon
emissions related to
energy use in homes by
x%)
Mechanisms: What
specific types of
instruments are utilized?
(e.g. amount of grants
and loans, savings target
defined in Supplier
Obligation, requirements
of building regulations)
(e.g. Supplier Obligation,
grant and loans scheme,
building regulations)
Calibrations: What are
the specific ways in
which the instrument is
used?
(e.g. publication of
building regulations and
guidelines)
Source: Modified from Howlett and Cashore (2007; 2009)
For the purpose of this thesis the Howlett and Cashore taxonomy of ‘policy’ will be used
in the empirical research. While this thesis does not look at all the elements of policy when
61
Methodology
analysing policy change it will refer to the taxonomy to provide clarity as to which element
is meant. Policy change is defined as a modification of the elements of policy as defined
above. Hence policy change is not simply the modification of existing policies but also the
creation of new policies. Change can be measured by looking at individual elements over
time. While some change is quantitative in nature (e.g. doubling the target for policy x,
increasing the budget for a certain policy), other types of change can only be described
qualitatively (e.g. change of policy goals and instrument logic).
When determining the unit of analysis the following rationale was adhered to: First, the
unit of analysis had to display long-term policy change with regard to the central research
question of this thesis. Second, the interest of this thesis lies with home energy efficiency
policy and the unit of analysis had to capture key features of this policy area. Third, the
unit of analysis had to be suitable to carry out systematic process analysis. It would be a
herculean task to carry out systematic process analysis for all the home energy efficiency
policies in the UK and Germany. Therefore only one policy instrument per country could
realistically be analysed. By making this choice, this thesis focuses on both ends and
means of specific on-the-ground measures (shaded column in Table 2).
The policy instruments chosen represent some of the most important features of home
energy efficiency policy. In the UK, the Supplier Obligation (SO) is the most important
instrument to deliver energy and carbon savings in the domestic sector. A review of the
first Energy Efficiency Commitment scheme, EEC 1, describes the SO as ‘the
overnment’s key energy efficiency policy for existing households’ (OFGEM 2005). Both
the 2004 and 2007 Energy Efficiency Action Plan highlight the SO as the principal policy
mechanism to deliver energy savings in the domestic sector (2004b, 2007b). As outlined in
the Low Carbon Transition Plan, future GHG emission reductions in the domestic sector
62
Methodology
are assumed to come mainly from an extension of the SO with an increasing target going
forward (DECC 2009c).
In Germany, the KfW loans and grants schemes for energy efficiency measures are the
most important policy instruments for saving all energy in the domestic sector. For
delivering energy efficiency in the existing housing stock the Integrated Climate
Programme highlights the CO2 Buildings Rehabilitation Programme as the principal policy
instrument (BMU 2007). Also Eichhammer et al. (2006) stress that German home energy
efficiency policy is dominated by the KfW schemes.
Therefore the unit of analysis in the UK is the Supplier Obligation and in Germany it is the
CBRP.
3.5 Research methods
This thesis uses a mix of qualitative methods as suggested for studies including systematic
process analysis (George and Bennett 2005). Quantitative methods play an important role
when investigating relationships of different factors statistically. However, when trying to
uncover causal mechanisms, which this thesis is aiming to do, qualitative research methods
are the only approach that can be taken in a meaningful way ( l ser and Laudel 2004).
Qualitative research methods such as semi-structured interviews and document analysis
have been criticised for not being sufficiently rigorous. Some of this critique is relevant for
this thesis and will be considered in the following.
3.5.1 Critical assessment of qualitative research methods
Diefenbach (2009) presents a list of 16 potential methodological problems raised by
critics. The most relevant issues with general regard to qualitative research put forward by
Diefenbach are briefly discussed below while more specific problems of doing interviews
are addressed in the relevant section on interviews.
63
Methodology
First, the research question may be biased due to the implicit assumptions, worldviews,
preconceptions, and single-mindedness of the researcher (Collins 1992). While
quantitative research suffers from similar problems, one way of dealing with these issues
in qualitative research is to make the position of the researcher explicit (Pyett 2003).
Second, the research question may not be clearly defined at the beginning of the
investigation and is likely to change over the course of the research (Maso 1989).
However, this is not necessarily a weakness of qualitative research and can even be a
strength. As Diefenbach (2009, p. 877) puts it, ‘the (re-) formulation of the research
question (or adding new ones) is a sign for progress, for an increasingly better and deeper
knowledge and understanding of the objects of reasoning and recognition of emerging
patterns’. Third, the research methods used in qualitative research are not as precise and
clear-cut as quantitative methods (Whittemore 2001). Too often qualitative researchers do
not state explicitly the methods that have been used and in which manner. This thesis
makes the methods used explicit and highlights any methodical problems and areas of
potential bias. Fifth, interpretation of results is biased by the researcher’s subjectivity.
While this is certainly the case one way of taking a critical perspective is to use a range of
different potential explanations and test their explanatory power. Finally, qualitative
research may produce merely an endless number of case studies that do not advance social
science and theory development. While this certainly may be a problem of certain types of
research, systematic process analysis type research is very much theory led and aims at
testing as well as reformulating existing theories (George and Bennett 2005). Qualitative
research can go beyond simple testing of cause and effect and uncover the causal
mechanisms that link (or does not link) an effect to a cause. However, this thesis also
makes use of quantitative data where appropriate (e.g. energy price data). Such an
64
Methodology
approach can help to support claims made based on qualitative research and provide
additional evidence ( l ser and Laudel
).
Further methodological problems specific to the methods used in this thesis are discussed
below. The two main methods used in systematic process analysis are interviews and
document analysis:
‘In process tracing, the researcher examines histories, archival documents, interview
transcripts, and other sources to see whether the causal process a theory hypothesizes or
implies in a case is in fact evident in the sequence and values of the intervening variables in
that case. (George and Bennett 2005, p. 6)
Following
eorge and Bennett’s suggestion, both document analysis and interviews are
used in this research. Both are critically discussed in more detail below.
3.5.2 Document analysis
To begin with, published documents will be analysed and the long-term change of the
policies under investigation in this thesis will be reconstructed as far as possible.
Documents that will be used in the analysis include relevant academic literature, various
publications by government departments such as DECC, BERR, DEFRA, BMU and
BMVBS, regulators’ publications such as the annual reports issued by OF EM on the
Supplier Obligation, the reports commissioned by KfW on the German programmes for
building rehabilitation, party manifestos, speeches by politicians and stakeholders,
parliamentary hearings, newspaper articles, and other literature offering historical accounts
of domestic energy efficiency policy.
However, documents often do not contain all the information required. For example,
official documents often only contain limited information on an issue because certain
aspects of information were not considered as being important. Documents may also get
65
Methodology
lost, although this is less of a problem for research dealing with fairly recent policies.
Some documents may also be withheld from public analysis for secrecy reasons. If
documents are available, they may only contain an incomplete or distorted account of the
events in question. Government documents, for example, may present issues in a manner
that implies consensus and agreement while in fact a lot of decisions in policy making are
characterised by conflicts and disagreements. In some cases there are simply too many
documents to analyse making it difficult to identify the most relevant ones (Tansey 2007).
Document analysis is often used together with other research methods as a means of
triangulation i.e. ‘the combination of methodologies in the study of the same phenomenon’
(Denzin 1970, p. 291). Using it in combination with qualitative interviews is common
practice in social science research (Bowen 2009). As Kramer et al. (1990, p. 214) put it,
‘documents can enrich, and be enriched by, oral history’. Hence interviews can
complement document analysis and help address some of the problems described above.
3.5.3 Interviews
George and Bennett (2005) outline various purposes of interviews in systematic process
analysis. First, interviews can be a means of gathering basic information about a case.
Some information may simply not be available in published documents or the researcher
may have been unable to identify the relevant documents. Second, interviews can fill in
gaps in existing historical accounts. The documents available may not provide all the
information required and interviews can help with providing missing bits of information.
Interviews can help ‘to establish the decisions and actions that lay behind an event or
series of events’ (Tansey, p. 766-767). Respondents’ testimonies can be used to reconstruct
political episodes ‘stitching together various accounts to form a broader picture of a
complex phenomenon’ (p. 767). Third, interviews may serve theoretical purposes such as
finding omitted variables and hypotheses. As outlined above, this is one of the advantages
66
Methodology
of systematic process analysis. Finally, interviews can help the researcher to uncover
causal processes. For example, high energy prices may go hand in hand with new policy
initiatives in home energy efficiency but in order to understand how high prices translated
into policy change interviews can provide valuable information.
The people interviewed are experts in the area of the policies investigated in this study.
However, research aiming to reconstruct past political processes does not analyse the
experts as the main object of investigation (e.g. their backgrounds etc.) but makes use of
their knowledge of these processes putting the processes at the centre of the research
( l ser and Laudel
). An expert is a person who had or has responsibilities for areas
relevant to the case investigated or who has privileged access to specialist knowledge. This
person often represents an organisation without being necessarily at the top of the
hierarchy in an organisation but he or she may as well be part of the middle management
or an even lower level (Meuser and Nagel 1991).
In total, 26 interviews were carried out as part of this research project. Note that in
Germany members of parliament were interviewed whereas in the UK this was not the
case. The reason for this lies in the different responsibilities of members of parliament in
both countries: In Germany, the budget and other features of the CBRP require approval
and go through a complex parliamentary process as outlined in Section 7.1.2. As part of
this process members of parliament decide on any modifications the CBRP is subject to. In
Britain, however, the responsible government department sets out the secondary legislation
which specifies the design features of the SO. Parliament can only outright reject or accept
the proposed secondary legislation but rejection happens very rarely - the last time a piece
of secondary legislation was not approved by the House of Commons was more than 40
years ago (House of Commons Information Office 2008). Because the parliamentary
67
Methodology
process is merely a formality with no significance for the development of the SO no
members of parliament were interviewed in Britain.
Drawbacks of interviews
However, as with any research method, there are certain drawbacks to interviewing.
Interviewees may present false or misleading information because of their particular
interests (Diefenbach 2009). For example, if interviewing politicians, interviewees may
slant their accounts and provide a misleading picture of their role in a particular event
(Kramer et al. 1990). They may overstate or downplay certain factors, frame issues in a
manner that is favourable to them and the organisation the interviewee worked for at the
time of the event in question. These issues may be even more important if the interviewee
is working for an organisation at the time of the interview. The interviewee may merely
present an official position of that organisation rather than providing the sort of insights
that are required to answer the research questions. The background of an interviewee is
likely to have a major impact on the way issues are remembered and framed. Furthermore,
if the interview takes place years or even decades after the event under investigation,
lapses of memory may lead to partial and even misleading information (Kramer et al.
1990). Hence there is no such thing as truly objective information that can simply be
extracted in interviews.
However, this does not mean that information gathered in interviews cannot provide
important insights. George and Bennett (2005) provide a framework that can help the
researcher to assess the evidentiary value of the information gathered. They suggest that
one should ask who is speaking, to whom, for what purpose, and under what
circumstances. Furthermore, interview results can be cross-checked with other sources
(Diefenbach 2009), for example when documents exist that contain similar information.
68
Methodology
Sampling
It is also crucial to identify the right people when doing interviews in the context of
systematic process analysis. While many interviewing techniques include random
sampling, systematic process analysis requires that only those who have in-depth
knowledge about very specific events and issues are interviewed. In contrast to random
sampling, no clear-cut and well-developed concepts are available for non-probabilistic
sampling (Wengraf 2001). When selecting interviewees, the researcher can use different
sampling techniques. First, he or she may choose people because it is convenient to do so.
Existing contacts can serve as a pool of potential interviewees. However, the danger of this
approach is that important people may be missed. Second, quota sampling may be used
where representatives of each group are interviewed according to the proportion of this
group to the total number of actors. This approach requires detailed knowledge about the
population of actors and may not always be feasible. Third, in an approach using purposive
sampling, only those people who are deemed most appropriate in the context of the case
study and research question are interviewed. The problem with purposive sampling is that
the researcher may not identify key interviewees either because of a lack of information or
because the selection is biased in a one-directional way. Finally, snowball or chain-referral
sampling can be used where individuals are identified via suggestions by other
interviewees. However, because interviewees may tend to suggest other people who share
similar views, the results obtained may be skewed in one particular direction (Tansey
2007).
Another issue is the number of interviews carried out; qualitative researchers have been
criticised for relying on too few interviews. Yet, there are trade-offs between depth and
breadth and there is no perfect number of interviewees. It is up to the researcher to
69
Methodology
determine how many interviews are needed to answer the research questions (Diefenbach
2009). In sum, there is no perfect non-probabilistic sampling approach.
To avoid some of the problems of non-probabilistic sampling, the approach chosen for this
thesis is a mix of purposive sampling and snowball sampling. The aim is to identify
interviewees who fulfil the criteria for interviewee selection presented by Tansey (2007):
they have been significant players, they have good memory of the events, and they are
willing to disclose their knowledge of events in an impartial manner. Prior knowledge of
the case and the document analysis will be used to develop an initial list of potential
interviewees deemed most appropriate i.e. those who are likely to be able to provide the
information needed in order to answer the research questions. Thereby a critical number of
people will be reached to begin with. The list of people identified during purposive
sampling will be crosschecked with experts in those areas to ensure good coverage. The
initial list of potential interviewees will then be used to identify further interviewees by
doing what is called reputational snowball sampling (Goodman 1961). This method can be
particularly useful when trying to identify key policy makers (Farquharson 2005). The
approach outlined above, i.e. purposive sampling in combination with snowball sampling,
has been suggested for case study research using systematic process analysis (Tansey
2007).
Semi-structured interviews
In light of the above, semi-structured interview techniques are the most appropriate form
of interviews for systematic process analysis (Tansey 2007). Expert interviews should be
semi-structured following an interview guide when very specific information is required
and a range of topics needs to be covered. The reason for this is that a non-structured
interview, such as is used in narrative interviews, is unlikely to cover all the aspects of the
70
Methodology
case investigated when trying to reconstruct past events and processes. A fully structured
interview on the other hand might exclude certain issues from the start because only those
points can be covered that are part of the questionnaire ( l ser and Laudel 2004).
Semi-structured interviews require a number of initial questions to be prepared but
probably more than half of responses to what the interviewees say may be improvised
during the interview. This does not mean that less preparation is necessary – to the
contrary, the interviewer needs to be extremely well prepared to be able to ask the right
questions spontaneously. Good knowledge of both the theories used and the case study
area is an obligatory prerequisite of a successful interview (Wengraf 2001).
Following Wengraf’s suggestion, interview questions are derived from the theory
questions presented above, are tailored towards the type of interviewee, and presented in a
way that is as clear as possible to the interviewee. Questions should ideally echo the theory
questions without revealing too much about them. Also, interview questions should not be
suggestive or framed in a way that is favourable to one’s preferred hypotheses (Kvale
2007; Wengraf 2001).
Pilot study
It is generally recommended to test the interview questionnaire in a short pilot study
(Wengraf 2001). If there is only a limited number of people that can be interviewed, for
example when only a very few persons were involved in a particular process or only a
small number of people understand the issue of concern sufficiently well, pilot studies can
be crucial for avoiding unsatisfactory results ( l ser and Laudel
). Testing the
questionnaire with someone who is not one of the key interviewees but still has enough
expertise in the area can eliminate questions that are unclear or misleading.
71
Methodology
Prior to conducting the interviews the list of questions used for the purpose of this thesis
was tested with one pilot interviewee. Following the interview the list of questions was
modified taking into account the feedback by the interviewee and the ability of the initial
questions to generate the data needed.
3.5.4 Qualitative content analysis
Qualitative content analysis is an excellent way of analysing interviews when investigating
causal mechanisms ( l ser and Laudel
). This section provides an overview of
qualitative content analysis in general and the specific approach chosen for this study.
Overview
Qualitative content analysis is ‘the use of replicable and valid method for making specific
inferences from text to other states or properties of its source’ (Krippendorff 1969, p. 103).
It is particularly useful for coping with large amounts of data and helps analysing it in a
structured manner (Krippendorff 2004). According to Babbie, qualitative content analysis
is ‘essentially a coding operation’ (2001, p. 309). Coding is ‘the process of transforming
raw data into a standardized form’ (ibid.) making it possible to analyse a text in a
systematic manner.
The transcripts of the interviews are analysed step-by-step following clearly defined
procedures devising the material into content analytical units. Based on the research
question the different elements of the text are categorised. A category is ‘a group of words
with similar meaning or connotations’ (Weber 1990, p. 37). The categories are revised in
the process and there may be several feedback loops in the process. Qualitative content
analysis is led by the criteria of reliability and validity. The process of analysis should be
inter-subjectively comprehensible so the results can be compared to other studies and to
carry out checks of reliability (Mayring 2000). This can be ensured by describing the
72
Methodology
process of analysis in detail uncovering how the categories have been constructed and how
inferences have been made. Authentic citations also help increasing the trustworthiness of
the research (Elo and Kyngäs 2008).
There is no clearly defined way of doing qualitative content analysis as it is very case
specific and depends very much on the type of investigation. However, there are two
approaches in general that can be followed: First, an inductive approach may be chosen
when the different categories cannot be specified prior to the analysis. Inductive content
analysis is a reiterative process whereby the researcher identifies categories based on the
text, groups and organises those categories, and finally analyses the text with a final list of
categories. Second, a deductive approach is more suitable when a well-defined theory that
can be operationalised in variables to be tested (Elo and Kyngäs 2008; Mayring 2000).
Drawbacks
Similar to other methods, there are limitations and drawbacks to qualitative content
analysis. According to Mayring (2000), qualitative content analysis is less suitable in cases
where the research question is highly open-ended, explorative, variable and where working
with categories would be a restriction. In the case of this study, however, a theory-guided
step-by-step approach is chosen making qualitative content analysis a suitable means of
deriving results.
Qualitative content analysis has also been criticised for being susceptible to the effects of
researcher biases (Kolbe and Burnett 1991). Reliability problems can occur due to the
‘ambiguity of word meanings, category definitions, or other coding rules’ (Weber 1990, p.
15). Because the analysing process requires a significant amount of interpretation by the
researcher it is important to critically reflect the context in which the interviews were
conducted. According to Dey (1993), when reading the data the questions one has to bear
73
Methodology
in mind are: Who is telling? Where is this happening? When did it happen? What is
happening? Why? The reliability problem can be addressed by being explicit about the
coding process and by developing a set of clear recording instructions (Stemler 2001).
Furthermore, there may be problems with the validity of the results of qualitative content
analysis. When it comes to the reconstruction of historical processes a single person’s
claims cannot always be relied upon. Hence inferences made on the basis of interview data
need to be validated by using multiple sources and by triangulation of methods (Erlandson
1993).
Approach of this study
This thesis followed the approach by Elo and Kyngäs (2008) who propose to use inductive
coding when preliminary knowledge is limited or fragmented, which was the case in this
research project. In a first stage of open coding the interview transcripts were coded freely
without predefined categories. The software package NVivo was used for the coding
process. In order to limit the number of categories, open coding was followed by grouping
the categories identified depending on whether they were similar or dissimilar. Finally, in
an abstraction process more general conceptual categories were developed which could
then be used to refine the conceptual framework and evaluate the hypotheses.
3.6 Operationalisation
In order to operationalise the conceptual framework and its components, this thesis follows
Peter Hall’s advice on the use of systematic process analysis for the purpose of theoryoriented explanations. This approach ‘attaches less value to securing precise parameter
estimates for a few key variables seen as the ‘ultimate causes’ of the outcome and more
value to identifying regularities in the causal chain through which the relevant outcome is
74
Methodology
generated’ (Hall 2008, p. 3). Rather, the ‘focus is on elucidating the process whereby the
relevant variables have effects [my emphasis]’ (ibid.).
Key processes and mechanisms of the conceptual framework developed in section 2
include pressures external to the subsystem, friction, accumulation of pressures, positive
and negative policy feedback processes.
External pressures manifest themselves in the shape of information that is provided by the
media, lobby groups (including NGOs), experts, and government agencies. For example,
climate change research generates scientific reports. These reports are then analysed with
regard to policy implications in structured (for example through the IPCC reporting
procedures including summaries for policy makers) or unstructured ways (for example
NGOs using new academic studies to make a case for stricter climate policy). Other
pressures may be more immediate and can be experienced by the wider public directly.
Rising energy prices, for instance, regularly lead to discussions about the appropriate
policy response.
The conceptual framework assumes that external pressures accumulate due to the limited
capability of the system to deal with these pressures. Pressure accumulation implies that
over time the stress on the system continues (or even increases) because the measures
taken to ease the pressure are not sufficient. For example, if energy prices continue to rise
and government programmes on fuel poverty are unambitious fuel poverty is likely to
remain at high levels or increase. This would not be the case if the policy system addressed
the problem of rising energy prices and low incomes comprehensively. Furthermore,
pressure accumulation is likely to be highlighted by the media, NGOs, and interest groups
pointing out the inability of existing policies to deal with the problems created by external
pressures. Finally, the accumulation of pressures may also show in falling behind on
meeting policy objectives.
75
Methodology
The concept of friction is based on two causal mechanisms: First, the limited cognitive
ability of actors and institutions to process information and translate policy inputs in policy
outputs. Second, the limited capability of institutions to deal with policy inputs due to
institutional structures and legacies. While it is difficult to observe friction directly, there
are indirect indicators. One can assume that the constrained administrative capability of
institutions responsible for policy instrument implementation may be highlighted at some
point by media, government, experts, interest groups, and scientists. In this case an
investigation of the reasons for the constraints can unveil the causal mechanism such as
rules not allowing the institution in question to act more swiftly or limited staff to deal
with the issues. Also, friction is expected to result in certain issues being more prevalent
whereas others are ignored.
The third mechanism of the conceptual framework is policy feedback. If positive feedback
processes are at work, these can show in a number of ways. For example, if the policy
instrument is supported by various actors it is likely that these actors point out the positive
aspects of the existing policy instrument in order to maintain or expand the existing policy.
The existence and expansion of interest groups that benefit from an existing policy
instrument form another indicator for positive policy feedback. The same should be the
case for the contraction of interest groups that do not benefit from the existing policy
instrument. Increasing administrative capability is another type of positive policy feedback
and may show up in increasing confidence with the operation of the existing policy
instrument expressed in official documents or interviews. Also, if the existing policy
instrument has positive fiscal effects highlighted by various actors it is likely that positive
feedback mechanisms are at work. Finally, compatibility of the existing policy instrument
with the prevailing ideational legacy generates positive feedback effects i.e. the higher the
ideational fit, the more likely it is that the policy instrument will prevail. The ideational fit
76
Methodology
may show in government documents, interviews, parliamentary hearings etc. stating a
preference for particular types of policy instruments (e.g. market based or command and
control).
Following this logic, negative feedback comprises the respective opposite indicators:
interest group opposition to an existing policy instrument, existence and expansion of
interest groups that are disadvantaged by and contraction of interest groups that benefit
from the existing policy instrument, diminishing confidence with operation of an existing
policy instrument due to worsened administrative capabilities, negative fiscal effects of an
existing policy instrument, ideational misfit.
Ultimately, this thesis aims to explain policy change focusing on policy instruments (see
section 3.4.3). Thus any changes in the calibration of the policy instruments analysed
indicate policy change. Changes may relate to the overall target of the policy instruments,
the funding levels, technical specifications, eligibility criteria, and other aspects.
The following summary table presents indicators and sources of information for these
processes and mechanisms.
77
Methodology
Process /
Mechanism
Indicator
Sources of information
Pressure external to
subsystem
Issues raised by media, government, experts,
interest groups, scientists
Government documents
Parliamentary hearings
Commissioned reports
Media articles
Secondary sources
Statistics
Accumulation of
pressures
Continued pressure indicated as above
Inability of existing policy to deal with
pressures highlighted by media, government,
experts, interest groups, scientists
Existing policy instrument falls behind its
initial objectives
Interview transcripts
Media articles
Secondary sources
Government documents
Parliamentary hearings
Commissioned reports
Statistics
Friction
Constrained administrative capability of
institutions responsible for policy instrument
implementation highlighted by media,
government, experts, interest groups,
scientists
Prevalence of other issues and ignorance of
pressures
Interview transcripts
Media articles
Secondary sources
Positive policy
feedback
Positive aspects of existing policy instrument
highlighted by various actors
Existence and expansion of interest groups
that benefit from existing policy instrument
Contraction of interest groups that do not
benefit from existing policy instrument
Increasing confidence with operation of
existing policy instrument due to improved
administrative capabilities
Positive fiscal effects of existing policy
instrument
Existing policy instrument fits with
prevailing ideational legacy
Interview transcripts
Consultation responses
Evaluations
Statistics
Secondary sources
Government documents
Parliamentary hearings
Negative policy
feedback
Negative aspects of existing policy
instrument highlighted by various actors
Contraction of interest groups that benefit
from existing policy instrument
Existence and expansion of interest groups
that are disadvantaged by existing policy
instrument
Diminishing confidence with operation of
existing policy instrument due to worsened
administrative capabilities
Negative fiscal effects of existing policy
instrument
Existing policy instrument does not fit with
prevailing ideational legacy
Interview transcripts
Consultation responses
Secondary sources
Government documents
Parliamentary hearings
Statistics
78
Methodology
Process /
Mechanism
Indicator
Sources of information
Policy change
Changes in the setting of the policy
instrument including overall targets, funding
levels, technical specifications, sub-targets,
eligibility criteria etc.
Government documents
Conclusions
In light of the breadth of potential methodological approaches in political science, this
chapter presented and justified the methods and the research design employed. At a more
general level, the chapter positioned the research project in ontological and
epistemological terms determined by the theories used and the conceptual framework
developed in chapter 2. With regard to ontology, this thesis attributes particular
explanatory power to institutions, actors, time and the sequence of events.
Epistemologically, the thesis is firmly rooted within the critical realist perspective
acknowledging the limitations of theories and methods to capture reality sufficiently.
Following the ontological and epistemological positioning, the chapter continued with the
introduction and justification of the case study method discussing its potential as well as its
limitations. It argued for a particular (qualitative) case study method, namely systematic
process analysis developed by Peter Hall, because it is particularly suitable for unveiling
causal mechanisms. After elaborating on systematic process analysis, the chapter provided
the rationale for the case selection including the countries chosen, the time frame, and the
unit of analysis. In a research methods section the chapter critically discussed the validity
of qualitative methods in general and introduced the methods used in this thesis including
document analysis, semi-structured interviews, and qualitative content analysis. Finally,
the chapter concluded by operationalising the conceptual framework identifying indicators
and potential sources of information. In the next chapter, the methodology discussed will
be applied to the first case study, the Supplier Obligation in the UK.
79
Case I: Supplier Obligation in the UK
4 Case I: Supplier Obligation in the UK
Of course, one should bring order into history […]. Every science is, among other things, a
method of ordering, simplifying, making the indigestible digestible for the mind. […] To study
history means recognising and accepting chaos, and nevertheless retaining faith in order and
meaning.
Hermann Hesse (1972)
‘[M]aking the indigestible digestible for the mind’, as Hermann Hesse put it, is the task of
this chapter. Of course, describing and analysing the 16 years of history of the Supplier
Obligation (SO) within one chapter cannot do justice to all its aspects. Hesse’s quote
suggests that ordering comes at the cost of simplifying. Fully aware of this, I focus on key
issues and changes rather than aiming to provide an analysis covering every detail. While
much of the following is based on my interpretation of the course of events, I draw on 14
interviews with experts who observed the SO’s history either as analysts or as
practitioners. In addition, I look at previous academic work. Hence, my analysis is heavily
indebted to other people’s interpretations.
The chapter consists of two distinct parts, one is descriptive and the other analytical.
Before an in-depth analysis of policy change and the SO is carried out, background
information on the SO and the main changes is provided in a descriptive section.
4.1 Background and main changes of the Supplier Obligation
This part of the thesis looks in detail at the past development of the principal policy
instrument in Great Britain in the area of home energy efficiency, the Supplier Obligation
(SO). First, the basic architecture of the SO is introduced. Second, the main changes that
80
Case I: Supplier Obligation in the UK
happened since its inception in 1994 are laid out. This section is then followed by a
detailed analysis of the drivers and patterns of change from 1994-2010.
4.1.1 Basic architecture of the SO
In 1994, the government introduced an obligation on electricity suppliers to deliver a
certain amount of energy savings at the customer end. Although the initial obligations
were set at a low level and covered only electricity suppliers, the instrument became one of
the key features of UK energy efficiency policy eventually covering all suppliers with
more than 250,000 gas and/ or electricity customers in Great Britain. Northern Ireland is
not covered by the SO.
The basic concept of the SO is that Government imposes a savings target on energy
companies that has to be achieved at the customer end. This target may relate to all energy
consumption or carbon emissions. Businesses and industrial end-users are not covered by
the scheme; they are targeted by other policy instruments such as the Climate Change
Levy and Climate Change Agreements, as well as the recently introduced Carbon
Reduction Commitment. The target is set by the Department of Energy and Climate
Change (DECC) for a defined period of time using a bottom-up approach assuming an
illustrative mix of various energy saving measures that is likely to be used in order to
deliver the obligation. Note that it is established in comparison to a baseline and does not
require a reduction of final all energy use. Rather, it is defined in terms of lifetime savings
achieved by the measures likely to be promoted by the energy suppliers. The energy
regulator, the Office of Gas and Electricity Markets (OFGEM), is responsible for
administering the SO and enforcing it. OFGEM defines individual savings targets for each
energy supplier and checks whether the obligated companies achieve them. There are
various routes the energy suppliers use (Figure 3): First, the energy suppliers contract
81
Case I: Supplier Obligation in the UK
installers of energy saving measures (for example cavity wall insulation) that carry out the
work in homes according to a defined standard and with a certain benchmark for energy
and/or carbon savings. Second, they may subsidise energy efficient products (for example
insulation material) sold via ‘do it yourself’ (DIY) and other retailers. Third, energy
suppliers deliver their obligation through work with managing agents, Local Authorities,
supermarkets and other retailers, which in turn subcontract installers and manage the
delivery process. Finally, energy companies may choose to work with the home occupants
directly. In the past, energy companies have for example promoted the use of compact
fluorescent lamps (CFLs) via mass mail-outs of free light bulbs, although this is now
prohibited. More recently, some energy companies, such as British Gas, set up their own
insulation business and deliver most of their obligation themselves. In theory, the energy
suppliers can trade their obligations amongst themselves. In practice, this has not happened
on a large scale. Suppliers can combine the various delivery routes and may choose to
focus their efforts on one particular route or spread their activities.
The SO is based on self-reporting and a certain degree of trust: While OFGEM review the
figures provided by the energy suppliers, they do not check measure by measure given the
administrative burden this would involve. In theory, energy suppliers could inflate their
figures, but the potential fine for doing this and for not achieving the target is significant:
up to 10% of global turnover of the obligated companies in case they miss their target.
82
Case I: Supplier Obligation in the UK
Figure 3: Mechanism of the Supplier Obligation in Great Britain
Source: Rosenow (2012)
While there was a succession of different SO schemes, the basic logic remained the same.
The first SO scheme was called Energy Efficiency Standards of Performance (EESoP) and
ran from 1994 to 1998. Its successors, EESoP 2 and EESoP 3, ran from 1998 to 2000 and
2000 to 2002, respectively. In
the scheme’s name was changed to Energy Efficiency
Commitment (EEC). EEC 1 was in place from 2002 to 2005 and EEC 2 from 2005 to
2008. EEC was eventually renamed in 2008 to the Carbon Emissions Reduction Target
(CERT) that operated from 2008 to 2012.
4.1.2 Main changes of SO from 1994 to 2010
As already indicated, the SO is characterised by a history of quite significant policy
change. This section only describes those changes without analysing the political dynamics
behind them – a detailed investigation of the policy processes behind the modifications
that occurred follows in a later section.
83
Case I: Supplier Obligation in the UK
The most important changes are to the elements listed in the following table:
Table 3: Main elements of policy change in case of the SO
Main elements of policy change
target size and changing target metric
costs of the SO
coverage
share and definition of the Priority Group
trading and carry-over
administering and target setting body
qualifying measures
monitoring and verification
Those changes will be explained in detail below.
Table 4 summarises how the scheme changed since its inception in 1994:
84
Case I: Supplier Obligation in the UK
Table 4: History of the SO in Great Britain
Name of scheme EESoP 1
EESoP 2
EESoP 3
EEC 1
EEC 2
CERT
Inception year
1994
1994-1998
1998
1998-2000
2000
2000-2002
2002
2002-2005
2005
2005-2008
2008
2008-2012
6.1 TWh
(lifetime)
2.7 TWh
(lifetime)
62 TWh
(lifetime)
130 TWh
(lifetime)
Implicit annual
target
Target group
1.5 TWh
(lifetime)
Public
Electricity
Suppliers
(PESs)
1.4 TWh
(lifetime)
Public
Electricity
Suppliers
(PESs)
4.9 TWh
Electricity & 6.1
TWh Gas
(lifetime)
5.5 TWh
(lifetime)
all licensed gas
and electricity
suppliers with
at least 50,000
domestic
customers
Coverage
Domestic and
small business
electricity
customers
Domestic and
small business
electricity
customers
Domestic and
small business
electricity and
gas customers
21 TWh
(lifetime)
all suppliers
with over
15,000 gas
and/or
electricity
domestic
customers
Domestic
electricity and
gas customers
43 TWh
(lifetime)
all suppliers
with over
50,000 gas
and/or
electricity
domestic
customers
Domestic
electricity and
gas customers
293 million t
CO2 (lifetime)
≍ 494 TWh
(lifetime)
~104 TWh
(lifetime)
all suppliers
with over
250,000 gas
and/or
electricity
domestic
customers
Domestic
electricity and
gas customers
Cost of
the programme
Annual
expenditure
Expenditure
allowance
£101.7 million
£48.1 million
£25 million
£24 million
£110 million
(indicative)
£55 million
£500 million
(indicative)
£167 million
£1,2 billion
(indicative)
£400 million
£5.6 billion
(indicative)
£1,158 million
£1 per
franchise
customer per
year
allowed through
the
supply price
control
£1 per
franchise
customer per
year
allowed through
the
supply price
restraint
£1.20 per
customer
per fuel per
annum,
indicative in
target
setting model
£3.60 per
customer per
fuel per annum,
indicative in
target setting
model
£9 per
customer per
fuel per annum,
indicative in
target setting
model
£51 per
customer per
annum,
indicative in
target setting
model
Allowance for
R&D and energy
monitoring
Percent of
savings in
priority group
3%
0.42%
0.5% for
monitoring and
0.25% for R&D
67% of
expenditure
(expected, not
compolsary)
none
none
requirement to
use variety of
measures
not applicable
not applicable
not applicable
50%
50%
40%
15% in Super
Priority Group
10%
yes
not precribed
unlimited
yes
not precribed
unlimited
yes
minimum levels
for some
measures (68%
from insulation)
Period*
Target
30% (expected, 65% of
not
expenditure
compulsory)
(expected, not
compulsory)
Carry over
none
none
Trading
none
none
Mix of measures requirement to requirement to
use variety of
use variety of
measures
measures
Administering
body
Target setting
body
OFFER
OFFER
OFGEM
OFGEM
OFGEM
OFGEM
OFFER
OFFER
OFGEM
DEFRA
DEFRA
DEFRA and then
DECC (since
October 2008)
* period of schemes from 01 April in start year to 31 March in end year; different for CERT where period from 01 April
in start year to 31 December in end year
Source: based on DECC (2010b, 2010c); DEFRA (2008); OFFER (1994, 1998a); OFGEM (2005, 2008c,
2009a, 2009c); OFGEM and Energy Saving Trust (2003)
In order to better understand the detailed changes that occurred, the different aspects of the
SO that changed over time are presented below.
85
Case I: Supplier Obligation in the UK
Target size
The target size of the SO increased significantly since 1994; the scale of change is really
quite remarkable. It should be noted, however, that there are some caveats to comparing
the different targets because of the different target metrics used.
The total energy saving target of the SO in 1994-1998 and 2008-2012 cannot be directly
compared because the EESoP 1-3 target was defined in terms of money to be spent
whereas EEC 1-2 defined the target in terms of TWh to be saved and CERT set the target
in Mt CO2. However, for the EESoP 1-3, the regulator published figures for the implicit
energy savings target, i.e. the expected savings generated by the expenditure. This can be
compared to the EEC 1-2 targets. According to OFGEM (2008d), the CERT target is
equivalent to doubling the target under EEC 2, which was 130 TWh. Taking into account
the 20% uplift in September 2008 and the extension of the scheme to December 2012, the
total (implicit) energy savings target of CERT is almost 500 TWh. Given that the implicit
EESoP 1 target was 6.1 TWh this means that the SO target increased eightyfold from
1994-1998 to 2008-2012.
However, these figures are not directly comparable either because the length of the
different schemes differs. Therefore, an average annual energy savings target is calculated
that allows direct comparison of the schemes (Figure 4).
86
Case I: Supplier Obligation in the UK
Figure 4: Annual energy savings target of SO in TWh
Source: based on DEFRA (2008); OFFER (1998a); OFGEM (2005, 2008c, 2009a, 2009c); OFGEM and
Energy Saving Trust (2003)
According to this calculation, the implicit annual energy savings targets increased almost
seventyfold from 1994-1998 to 2008-2012. However, for a number of reasons the targets
are not directly comparable. First, the initial target was set for electricity only and later
targets covered both gas and electricity customers. Second, after EESoP 3 the target metric
changed (see below) to fuel-weighted savings according to the carbon content. Third, the
schemes used different discount rates, which makes direct comparison difficult. While it
would be possible in theory to convert the targets to a more meaningful unit, it would not
change the overall picture significantly. The key point to be made is the overall magnitude
and pace of change rather than the precise target variation over time. Given the substantial
readjustment that would be required and the limited additional benefit of such a process,
no such conversion is undertaken as part of this thesis.
Target metric
As already indicated, the target metric changed several times:
87
Case I: Supplier Obligation in the UK

EESoP 1-3: EESoP 1-3 set the target in terms of money to be spent per customer.
Under EESoP 1 and 2, the target only related to electricity. EESoP 3 set a target for
both electricity and gas separately.

EEC 1-2: The EEC 1 and 2 targets were set in terms of fuel-standardised energy
savings, allowing suppliers to achieve savings in homes heated by gas, electricity, coal,
oil or LPG. Energy savings were carbon weighted and discounted in line with the HM
Treasury Green Book (guidelines for carrying out cost-benefit analysis), although the
rate changed over time.

CERT: CERT then changed the target from energy savings to carbon emissions and
abolished the discounting procedure (OFGEM 2009a).
The reasons for changing target metrics are manifold and will be explored in more detail
below.
Cost of programme
As a result of increasing targets, the cost of the programme to energy suppliers increased
from just £101.7m in EESoP 1 (£25 million per year) to £5.5b in CERT (£1.2 billion per
year). While EESoP 1 and 2 obliged energy suppliers to spend a certain amount of money,
later versions of the SO only provided indicative figures that were nonbinding.
In competitive markets suppliers in theory pass on the costs of the SO to their customers.
No real data is available to check whether this is actually the case (this is confidential
information not shared by energy suppliers) but it can be assumed that suppliers treat the
SO as any other cost imposed on their business, which will be reflected to some extent in
the energy bills.
88
Case I: Supplier Obligation in the UK
Whereas the costs of the SO were subject to supply price control (and the 1998 supply
price restraint) in earlier versions of the SO (EESoP 1 and 2) prescribing how much money
energy suppliers had to invest, expenditure in later versions did not fall under such tight
control and only indicative figures were provided by the responsible government
department. The average bill increased by only £1 per household per year during the early
EESoP schemes, but by more than £50 per household per year under CERT, meaning a
fiftyfold increase over 10 years.
However, it is difficult to estimate the exact increase in energy bills given that the above
figures are based on modelling exercises and that suppliers do not provide such kind of
information in a competitive market. Therefore, the reported figures on expenditure by the
energy suppliers have to be taken with a pinch of salt because since EESoP 3 they
represent the assumed expenditure. Research suggests that the actual expenditure levels for
EEC 1 were 20% lower (Lees 2006) and for EEC 2 were 23% lower than the figures
assumed by DEFRA (Lees 2008). However, even taking this into account the increase of
expenditure is remarkable and the impact on energy bills not insignificant.
Target group
The definition of companies falling under the obligation changed over time. At the
inception of the scheme, Public Electricity Suppliers (PESs) were the organisations
covered under the SO. PESs were the 14 companies created when the electricity market in
the United Kingdom was privatised in 1990. The individual targets for each PES were set
in relation to its number of customers. In 2000, as part of further restructuring of the
market under the Utilities Act 2000, the PESs were required to have separate licenses for
their supply business and distribution networks, which were renamed distribution network
operators (DNOs).
89
Case I: Supplier Obligation in the UK
EESoP 2 also covered PESs and targets were set using a similar methodology to EESoP 1.
However, greater regard was given to regional variations such as cost of supply (OFGEM
and Energy Saving Trust 2003). While EESoP 1-2 only covered electricity suppliers,
EESoP 3 put an obligation on all licensed electricity and gas suppliers with at least 50,000
domestic customers. As well as the 14 ex-PESs, four other suppliers were given targets.
EEC 1 set the limit at 15,000 domestic customers and twelve suppliers were covered.
OFGEM set progressively higher targets for suppliers with larger consumer numbers to
take into account any economies of scale that they were expected to achieve. Since EEC 2,
the threshold was raised again to 50,000 domestic customers. Due to consolidation in the
energy market, only eight suppliers were covered by the scheme. The trend of
consolidation continued and CERT covers just six suppliers with more than 250,000
customers.
Coverage
The coverage of customers changed as well. At first, energy savings had to be achieved at
both domestic and small business customers (EESoP 1-3). EESoP 1 covered domestic and
non-domestic customers with a maximum energy demand of 0.1 MW (OFFER 1994).
EESoP 2 had a reduced scope covering only households and non-domestic customers with
annual demand under 12,000 kWh (OFFER 1998a). EESoP 3 focused on domestic
customers, but allowed energy suppliers to raise funds by energy charges to small business
customers. Therefore, suppliers were permitted to provide some energy efficiency
assistance through their projects but not to design projects specifically for small businesses
(OFGEM 2000a). After EESoP 3 suppliers could only count savings in the domestic
sector.
90
Case I: Supplier Obligation in the UK
In the beginning, energy efficiency measures were restricted to suppliers’ own customers,
but eventually suppliers were able to target any household in Great Britain (since EESoP
3).
Share of Priority Group
CERT requires suppliers to achieve 40% of all savings in the so-called Priority Group
(PG), a group covering customers believed to be vulnerable and disadvantaged. This
requirement has not always been in place and earlier schemes did not set firm PG subtargets. EESoP 1 did not set a specific target for disadvantaged customers or those that
were in the PG of later schemes. OFFER argued that ‘it would not be helpful to set specific
quotas for savings to be achieved from projects covering [low income] customers’
(OFFER 1994, p. 7). However, when the Standards were set in 1994, OFFER expected
that 30% of the savings would be achieved in properties of low-income customers. For
EESoP 2, 65% of the total expenditure was expected to benefit low-income households
(OFFER 1998a). OFFER once again argued that it would not be appropriate to set a target
for low income households and that ‘each PES should take into account the needs of
different customer groups’ (OFFER 1998b, p. 10). For EESoP 3 OFGEM expected that
about two thirds of all projects would benefit low-income households, but did not set an
obligatory target either (OFGEM 2000a).
EEC 1 was the first scheme that put in place an obligatory target for the PG: 50% of all
savings had to be achieved within the PG (OFGEM 2001). The target did not change in
EEC 2 (OFGEM 2004a), but under CERT the target was reduced to 40% (OFGEM
2009c).
CERT also introduced a PG flexibility option allowing suppliers to reduce their PG target
from 40% to 35% if they spent equivalent funds on certain measures that would benefit a
91
Case I: Supplier Obligation in the UK
sub-group of the PG. The measures eligible were ground source heat pumps in respect of a
property that does not have a mains gas supply and solid wall insulation that lowers the Uvalue of the walls to 0.5W/m2K or less. Suppliers received an uplift of carbon savings if
they invested in those measures. The sub-group was defined as receiving benefits with
income below a threshold (£16,040 in 2009) (OFGEM 2009a).
Building on the flexibility mechanism, the CERT extension in June 2010 introduced an
additional Super PG (SPG). Suppliers were required to meet 15% of their total target
within an on-going 40% PG target in a subset of low income households that were
considered to be at high risk of fuel poverty (DECC 2010b).
Definition of PG
As outlined above, it was not before EEC 1 that a firm PG target was introduced.
However, EESoP 1 called on suppliers to take into account ‘in particular the interests of
those consumers who are elderly or disabled, who are in rural areas or who may have
difficulty in meeting their obligation to pay for electricity’ (OFFER 1994, p. 20). The same
definition of low income customers was used in EESoP 2 and 3 (OFFER 1998a; OFGEM
2000a). This definition of the group of disadvantaged customers is rooted in the 1986 Gas
Act and 1989 Electricity Act, which put a duty on the respective Director General of the
regulator to take into account the interests of those particular groups of customers.
An obligatory target for disadvantaged households was first introduced with EEC 1. The
PG under EEC 1 were defined as recipients of one of the following benefits: council tax
benefit; housing benefit; income support; an income-based jobseeker's allowance; an
attendance allowance; a disability living allowance; working families tax credit or disabled
persons tax credit; a war disablement pension which includes a mobility supplement or a
92
Case I: Supplier Obligation in the UK
constant attendance allowance; and a disablement pension which includes a constant
attendance allowance (OFGEM 2001).
EEC 2 followed the PG definition of EEC 1, but excluded people on disabled persons tax
credit and included people claiming child or working tax credits with income less than
£14,600 and people receiving state pension credits (OFGEM 2004a).
CERT’s P
definition was based on EEC
with slightly higher income limits for people
on child and working tax credits of £15,592. One important change was, however, the
inclusion of all people over 70 years old (OFGEM 2009a). As a result, PG activity in the
first year of CERT was focussed on the over 70s and not necessarily on those on low
incomes (OFGEM 2009c).
As already mentioned, the extension of CERT in 2010 also introduced a SPG target of
15%. The SPG focused on recipients of the following benefits:

Pension Credit;

Child Tax Credit under £16,190 income threshold; and

Income-based Job Seeker’s Allowance, Income-related Employment and Support
Allowance or Income support and one of the following:
- pensioner premium
- disability or severe disability premium
- award of child tax credit that also includes an element for a disabled, or severely
disabled, child or young person
- child under the age of five
The new SPG also replaced the previous sub-group definition of the PG flexibility
mechanism.
93
Case I: Supplier Obligation in the UK
Trading
Trading of energy saving obligations was first allowed in EEC 1. Suppliers were permitted
to buy certificates from or sell those to other suppliers. Under EEC 1, only two suppliers
were involved in trading obligations. Dee Valley traded its entire target with EDF Energy
(OFGEM 2005). During EEC 2, trading activity increased slightly, but remained at a low
level compared to the total share of the target. npower purchased PG savings from another
supplier, this contributed to nearly 2% of npower's target. Opus Energy traded its entire
target with Telecom Plus in order to comply with the obligation and Powergen met 1% of
its target by trading savings with another supplier (OFGEM 2008c). During CERT, British
Gas, SSE, ScottishPower and EDF traded a small share of their achieved carbon savings
with other suppliers. However, even though higher than during previous schemes, the total
share of traded carbon savings remained well below 5% (OFGEM 2012a).
Carry-over
Since EESoP 3 came into effect, suppliers were allowed to carry over energy savings from
one period of the SO to another; for example when suppliers had overachieved their targets
during EESoP 3, they could carry those savings over to EEC 1. In the beginning this was
limited to just 10% of the total target, but the cap was lifted with the inception of EEC 1.
About 3 TWh savings, equivalent to 5% of the EEC 1 target, were achieved by taking
forward savings from EESoP 3 (OFGEM 2005).
Six suppliers chose to carry over their excess measures from EEC 1 into EEC 2, equating
to roughly 25% of the EEC 2 target (OFGEM 2005). Therefore, energy suppliers were
ahead of the EEC 2 targets in 2005. Although the scheme had just started, EDF Energy had
already achieved more than half of its obligation for the next three years (ENDS 2005). In
94
Case I: Supplier Obligation in the UK
August 2006, suppliers had already met about 60% of the obligation, just over a year after
the start of EEC 2 (ENDS 2006d; OFGEM 2006).
Suppliers carried over excess savings from EEC 2 into CERT, which contributed to about
25% of the initial CERT target (OFGEM 2008c). Due to the CERT uplift in 2008 and the
extension until December 2012, the share of carry-over savings was just about 13%.
To sum it up, in the beginning of the SO (EESoP 1-2) suppliers were not allowed to carry
over savings from one scheme to another. However, from EESoP 3 onwards this option
was implemented and suppliers made use of this possibility.
Administering and target setting body
As part of the energy market liberalisation, the Gas Act 1986 and the Electricity Act 1989
gave powers to the Secretary of State to set up independent regulators OFGAS and
OFFER. OFFER set the energy saving targets and administered EESoP 1 and 2. OFGAS
agreed to extend EESoP to gas suppliers in 1999 but never administered EESoP 3 because
it was merged with OFFER resulting in the creation of OFGEM in 1999 (Bower 2003).
OFGEM subsequently set energy saving targets and administered EESoP 3.
A remarkable change to the distribution of responsibilities happened in 2000 when the
Utilities Act made provisions for energy saving targets to be set by DEFRA (following its
creation in 2008, DECC has been responsible for the target setting). The first target under
the new regime was set for EEC 1 in the Electricity and Gas (Energy Efficiency
Obligations) Order 2001. The Order also obliged OFGEM to administer EEC 1, which
included setting individual targets for suppliers, ensuring compliance, and reporting to the
Secretary of State. This structure was maintained for subsequent schemes.
95
Case I: Supplier Obligation in the UK
Throughout the different EESoP schemes, EST advised the regulator on how the standards
should be set and on what scale (OFGEM and Energy Saving Trust 2003). EST continued
to give advice to DEFRA and subsequently DECC on the SO. EST also ran some of the
energy savings projects in the early days of the SO on behalf of the energy suppliers.
Qualifying measures
Energy suppliers were always allowed to determine themselves the mix of measures used
to achieve their targets. There were, however, certain restrictions they had to follow. Under
EESoP 1-3 energy suppliers had to get approval from EST for the proposed mix of
measures and the regulator formally approved the proposed measures given EST’s
consent. Suppliers needed to demonstrate that they used a variety of measures in order to
achieve their saving obligations (OFFER 1994, 1998a; OFGEM 2000a).
After the restructuring of responsibilities, OFGEM was in charge of assessing the
measures that qualified for EEC 1 and its successors. OFGEM published lists of measures
that were deemed as qualifying actions (e.g. OFGEM 2002), but suppliers also had the
opportunity under EEC 1 to use new innovative measures not covered by the list. In order
to get approval from OFGEM for such measures, suppliers needed to provide independent
verification from a body accepted by OFGEM as being qualified to assess the savings
(OFGEM 2002).
EEC 1 introduced some restrictions concerning compact fluorescent lamps (CFLs) and
suppliers were only allowed to mail out four CFLs when provided for free or six when
customers contributed to the costs (OFGEM 2001). EEC 2 introduced further restrictions
and requirements regarding CFLs such as checking customer records to avoid multiple
CFL deliveries and to maximise the energy savings (OFGEM 2004a). Similar restrictions
were part of CERT, but CFLs were eventually banned from the scheme starting in January
96
Case I: Supplier Obligation in the UK
2010 only permitting CFLs provided via retail (OFGEM 2009c). With the extension of
CERT in June 2010, Government announced a total ban on CFLs under CERT (DECC
2010c). The main reason for this move was that incandescent bulbs are phased out under
EU legislation, more than 300 million CFLs had been distributed through the SO since
2002, and CFLs could not be seen as additional measures any longer (DECC 2010b).
As part of the CERT extension and for the first time since the inception of the SO, the
Government decided to set a minimum share for a particular type of measure. Suppliers
were required to achieve 68% of the CERT target by investing in insulation measures.
However, the minimum requirement only applied to the extension period and the
associated uplift of the total target was not applied retrospectively (DECC 2010b).
The mix of measures used by energy suppliers in order to achieve the required energy
savings changed over time (Figure 5). Insulation and lighting dominated the mix with an
increasing share of insulation since the inception of EESoP 2; lighting measures declined.
Earlier schemes (EESoP 1 and 2) had a 5% proportion of savings achieved in small
businesses, later schemes concentrated on domestic customers only. Because of the
inclusion of gas in EESoP 3, gas heating measures started to be part of the mix as well.
From EEC 1 onwards, suppliers were allowed to promote measures leading to savings of
any fuel used by households, not just electricity and gas. Note that not all measures
approved under the SO were energy efficiency measures. Energy suppliers used combined
heat and power energy generation (CHP) since the inception of the SO in 1994, although
the share remained very low. Later versions of the SO also saw some activity of microrenewables such as solar thermal, but so far this was, and still is, at a very low level
(mainly a result of higher costs). Also, both ground and air source heat pumps were used
97
Case I: Supplier Obligation in the UK
as part of the heating measures, but only contributed a small proportion to the total savings
achieved.
Figure 5: Mix of measures in different SO schemes
Source: based on OFGEM (2005, 2008c, 2010a); OFGEM and Energy Saving Trust (2003)
EEC 2 provided an incentive for energy suppliers to use innovative measures by increasing
the accounted energy savings of certain measures by 50%. The share of the innovation
measures of the total target was capped at 10%. All but one supplier used the incentive and
installed innovation measures such as solar thermal, CHP, ground and air source heat
pumps, highly energy efficient appliances, and window glazing (OFGEM 2008c).
CERT also contained provisions for innovative action and allowed up to 6% of a supplier’s
obligation to be met through them (8% where at least 2% was achieved by
microgeneration). The CERT amendments announced in September 2008 included an
increase of the cap to 10% (12% if 2% is met by microgeneration). In addition to the
measures which already fell under the EEC 2 specification of innovative action and which
CERT called ‘market transformation measures’, CERT allowed suppliers to try certain
measures classified as ‘demonstration actions’ (OFGEM 2009b). The 20% uplift of CERT
in 2008 also introduced the option to use behavioural measures such as real time displays
98
Case I: Supplier Obligation in the UK
(RTDs) and home energy advice packages (HEAs) (OFGEM 2009a). However,
behavioural measures are currently capped at % of a supplier’s target (OFGEM 2011a).
In summary, the main policy change that took place was the introduction of firm subtargets for insulation measures, the ban on CFLs, and the broadening of the eligible
measures. The majority of savings was related to insulation and lighting measures with an
increasing share of insulation measures over time. Other technologies included energy
efficiency appliances and heating systems, but their share declined over time. Only a
fraction of savings was achieved by promoting microgeneration/ CHP and behavioural
measures.
Monitoring and verification
Because the SO always focused on small energy users, first households and small
businesses and then just households, the approach taken to verify the savings was based on
deemed savings rather than measured savings. The values for those deemed savings were
generated by using models such as the BRE Domestic Energy Model (BREDEM).
Measures for which reliable engineering estimates of energy savings exist were given a
specific energy savings score (later carbon emissions savings) minus a reduction factor
which addressed rebound effects, free rider effects, and technical shortcomings of the
measures (for a detailed analysis of how this is done see chapter 6). However, suppliers
were required from the inception of the SO to undertake limited monitoring of the
measures installed in order to check that the energy savings assumed were in line with the
actual savings achieved. Particularly for more innovative and unproven measures this had
been the case.
During EESoP 1 to EESoP 3 5% of all insulation and heating measures had to be
monitored for quality and customer satisfaction by energy suppliers and an allowance was
99
Case I: Supplier Obligation in the UK
made for a proportion of suppliers’ expenditures to be allocated to fund monitoring
projects.
Similar requirements were in place for EEC 1. Suppliers had to monitor the quality of
installation of insulation or heating measures to a minimum of 5% of all measures; this did
not include monitoring all energy use. For DIY and innovative schemes, quality
monitoring requirements were determined by OFGEM on a scheme by scheme basis
(OFGEM 2001). Customer satisfaction needed to be monitored for all measures involving
CFLs but no specific procedures were prescribed. Suppliers were required to monitor 5%
of all DIY loft insulation measures with regard to customer satisfaction. For innovative
measures or other DIY schemes an appropriate monitoring methodology needed to be
agreed with OFGEM (OFGEM 2001).
Monitoring requirements became more complex with EEC 2 because of an increasing
number of eligible measures. Suppliers had to monitor the quality of 5% and customer
satisfaction of 1% of all professionally installed insulation measures as well as draught
proofing if combined with other insulation measures. The same requirements applied to
boilers, ground source heat pumps, solar water heating and fuel switching. A statistically
significant sample of DIY loft insulation and radiator panels had to be monitored and at
least 1% of all schemes had to be checked for consumer utilisation (i.e. how many actually
installed the insulation after purchase). Households that received CFLs free of charge were
monitored regarding consumer utilisation and energy suppliers had various options to
choose from to determine the sample size. Monitoring requirements of brown goods such
as set top boxes were decided on a case-by-case basis. CHP installations were monitored
by the provision of CHP Quality Assurance (CHPQA) certificates (OFGEM 2004a).
100
Case I: Supplier Obligation in the UK
CERT requirements for monitoring were built on EEC 2 and looked quite similar. Changes
include a cap at 1,000 customers on the requirement to monitor consumer utilisation for
CFLs. No technical monitoring was required for microgeneration measures that were
installed under the Microgeneration Certification Scheme. The CERT 20%-uplift
introduced real time displays (RTDs) and home energy advice packages (HEAs) as
additional measures qualifying for the scheme and required utilisation and evaluation
monitoring of 5% - or a statistically significant sample, whichever was smaller. Evaluation
monitoring was supposed to assess the effectiveness of the new measures (OFGEM
2009a).
4.2 Analysis of policy drivers
While the previous part was purely descriptive, the following delves deeper into the role of
the mechanisms that caused the changes described above. One interviewee stressed how
remarkably the SO changed over the years:
‘I don’t think anyone thought in 1993 ‘let’s make the electricity suppliers spend a couple of
million on energy saving projects’. I bet no one thought looking into the future when you had
£1.2 billion obligation on the energy suppliers. You start with that little acorn and then it just
snowballs and snowballs to where we’ve got to just following that route.’ (Interview UK12)
What this section does is to find answers to the question why the SO ‘snowballed and
snowballed’. While one could attempt to reconstruct the development of the SO
chronologically, an approach that clusters different episodes according to themes seems
more promising as a first step toward uncovering the causal mechanisms of change.
The following drivers of change have been identified:
101
Case I: Supplier Obligation in the UK
Table 5: Main policy drivers of the SO
Main drivers of policy change
market liberalisation
change of key personnel
transfer of target setting powers
rising energy prices
climate change policy
fuel poverty policy
growing energy efficiency industry with
increasing lobbying capability
perceived success of policy instrument
These mechanisms are not distinct but interrelated – for example the transfer of targetsetting powers from the regulator to the Secretary of State is directly related to the debates
that caused a change in personnel at OFGAS. Also, rising energy prices are a direct driver
of fuel poverty. However, for the purpose of this analysis separate treatment of these
mechanisms is more feasible and manageable due to the amount of information that is
processed. Note also that some of the drivers are more institutional (such as the transfer of
target setting powers) whereas others are more topical (such as climate change policy).
However, no clear distinction between the two categories can be made because all drivers
involve both topical as well as institutional aspects.
4.2.1 Market liberalisation
New energy paradigm
The beginnings of the SO can be found in the late 1980s and early 1990s. In the early
198 s, the new ‘neoliberal energy paradigm’ (Fudge et al. 2011) or the ‘new utility
regulatory regime’ (Helm 2002) began to take shape and the newly elected Conservative
government saw its task ‘to set a framework which will ensure that the market operates in
the energy sector with a minimum of distortion and energy is produced and consumed
102
Case I: Supplier Obligation in the UK
efficiently’ (Lawson 1982). So energy efficiency was very much a means to achieve
economic efficiency and maximise economic productivity of the economy at the time
(Interview UK1, UK2, UK3, UK10, UK13) and it was not debated in the context of
reducing carbon emissions as it is today (Interview UK13). The arrival of the new
conservative government was followed by initiatives to liberalise the markets for energy
and open them up for competition, as this was seen as the most suitable way of achieving
efficient supply and consumption of energy (Helm 2003). In 1986, British Gas was
privatised in the Gas Act 1986, followed by electricity in the Electricity Act 1989.
As a result of the energy market liberalisation in the 1980s two independent regulators
were created: the Office of Electricity Regulation (OFFER) and the Office of Gas Supply
(OFGAS). Their primary duty was to develop competitive markets and regulate prices in
those areas where competition was not feasible (e.g. natural monopolies) (Owen 2006).
However, the 1986
as Act also put a ‘duty to promote the efficient use of gas’ [my
emphasis] on OFGAS (HM Government 1986). The 1989 Electricity Act made provisions
for OFFER to ‘determine … standards of performance in connection with the promotion of
the efficient use of electricity by consumers as … ought to be achieved by … suppliers’
[my emphasis] (HM Government 1989). Whereas OFFER already had powers to set
standards of performance as laid out in the 1989 Electricity Act, the gas regulator was
eventually given powers to complement this duty by enabling standards of performance to
be set in the Competition and Service (Utilities) Act 1992.
Hence, the promotion of energy efficiency was enshrined in the primary legislation and
regulators had the task to ensure energy was supplied and consumed efficiently. There was
the assumption that once energy markets were fully liberalised energy efficiency services
would just be offered by the energy companies as part of their portfolio to compete against
103
Case I: Supplier Obligation in the UK
each other. However, until full competition was actually reached some sort of intervention
by the regulator to encourage greater efficiency on the demand side was deemed
acceptable. Therefore, a lot of the discussions at the time were framed in a language of
competition and economic efficiency:
You had to make a market regulation economic argument to the initial regulators. [...] It was
very much ‘we must get the market as efficient as possible’. So this was an argument that was
constructed that this was a substitute for very effective competition for the delivery of energy
efficiency services (Interview UK1).
Energy prices were regulated by using the RPI-X formula, which allowed companies to
increase their prices each year by an amount equal to the retail price index rate of inflation
(RPI) minus an efficiency percentage (X). Gas or electricity purchase costs could be
passed through 100% to customers. However, this was not the case for energy efficiency
measures (Owen 2006) and energy suppliers had to bear the cost without being able to
recoup the investment. Hence, some argued, here in the case of gas, that the RPI-X
formula was ‘an active disincentive operating on British as to undertake gas conservation
and efficiency investments which could provide the least-cost gas services to consumers’
(Brown 1990, p. 33).
There was, however, also the issue of limited public funding.
Public expenditure
In addition to the above, it was recognised that energy efficiency programmes would be
difficult to fund with public expenditure (Owen 1997) and the SO provided a mechanism
that generated funding through a premium on the energy bill (Interview UK1, UK2, UK3,
UK4, UK7, UK8, UK10, UK11, UK14).
104
Case I: Supplier Obligation in the UK
One interviewee argued that at the time ‘it was very clear sort of tactics by the
environmental movement in general to think about ways outside of public expenditure’
(Interview UK1). For those involved in developing ideas for new energy efficiency
policies ‘it was convenient in the sense of avoiding the constraints of public expenditure’
(Interview UK1 ) and a ‘covert way of making the market invest in public goods’
(Interview UK3) without consumers noticing who eventually pay for the policy (Interview
UK14).
Subsequently, the SO not only persisted, but also, as discussed earlier, increased
substantially in size and importance. One ‘reason it kept going in its form is that for
government it was quite convenient, it was kept off the public books’ (Interview UK8).
Hence there were two important issues at the time: First, the market liberalisation and the
existing quasi-monopolies of the utilities and second, the constraints of public expenditure
to fund energy efficiency programmes. Both issues provided the basis for utility-run
programmes rooted in ideas around LCP.
Least-Cost Planning
As a result of those debates in the 1980s, there was considerable interest in instruments to
incentivise energy efficiency and in particular in the so called Least-Cost-Planning
approach (LCP), which was used in the US for encouraging monopoly utilities to take into
account the demand side and potential energy savings options at the customer end
(Interview UK1, UK10, UK11). Because energy suppliers in the UK were privatised as
monopolies rather than in a competitive market, there was considerable interest in LCP as
a means of using utilities to promote energy efficiency (Eyre 2008).
105
Case I: Supplier Obligation in the UK
The concept of LCP was originally coined by the former director of the Office of
Conservation at the US Department of Energy, Roger Sant (1979). In short, LCP is based
on the assumption that people do not want to buy energy but energy services, and those
should be provided at lowest cost. Where it is cheaper to save a unit of energy rather than
supplying it, utilities should opt for the demand reduction option and vice versa. LCP was
heavily influenced by ideas postulated by Amory Lovins (1976, 1977), who advocated a
‘soft energy path’ utilising energy efficiency and renewable energy instead of just ramping
up the supply of fossil fuels and nuclear energy. Lovins later on coined the term
‘negawatts’, which he defined as a unit of energy avoided due to efficiency improvements
(Lovins 1985). It is easy to see why this particular policy instrument generated a lot of
interest in the UK as it fitted very well with the dominant way of thinking at the time, even
though it required some modification to fit the UK market (Eyre and Staniaszek 2005).
This idea of using utilities as the agents to roll out energy efficiency programmes was
discussed in the UK context. For example, in the mid-1980s, in a series of reports called
‘Lessons from America’, ACE presented the US way of approaching energy efficiency in
utility regulation. They conclude:
Perhaps the most vital lesson we can draw from these American programmes, and which has
equal validity in the UK as in the US, is the suitability of gas and electric utilities as vehicles
for promoting and delivering residential energy conservation programmes. (ACE 1985, p. 27)
From the beginning, ACE was a strong supporter of utility energy efficiency programmes
(Dawkins 1987). ACE can be classified as what policy change theorists call ‘policy
entrepreneurs’ who push new policy ideas (Kingdon 1984, 1995, 2002; Mintrom 1996).
The idea of using a LCP-type instrument in the UK was investigated further, for example
in a report by ACE commissioned by OFGAS, that generally deemed such an approach
106
Case I: Supplier Obligation in the UK
suitable and beneficial (Brown 1990). It also featured in an OFFER consultation on energy
efficiency (OFFER 1991), although it was argued in the document that LCP might not fit
with the UK context that well mainly because there was competition in supply. The then
Director General of OFFER, Steven Littlechild, supported the general principle of LCP,
but was sceptical about its applicability to the UK context because of a different industry
structure. Therefore, he said, ‘we should not rush into least-cost planning’ (House of
Commons Energy Committee 1991c).
Although there was some degree of scepticism around LCP, the general idea continued to
play an important role in the discussions around energy efficiency. The House of
Commons Energy Committee published a report in March 1991 calling for a reform of the
regulatory approach in order to incentivise utilities to undertake energy efficiency
measures following the LCP logic (House of Commons Energy Committee 1991a). Not
surprisingly, British Gas opposed such a move, arguing that the commercial pressures
already provided a strong incentive to promote the efficient use of gas, with or without
LCP, and that it would involve a lot of unnecessary bureaucracy (House of Commons
Energy Committee 1991b). Those concerns were repeated during an inquiry of the House
of Commons Environment Committee:
[...] putting an overall imposition on all utilities, gas, electric, such as reducing gas
consumption or electricity consumption by X per cent over Y years; such a target, or standard,
occasionally gets suggested. We believe that that kind of overriding, simplistic, almost
bureaucratic passing on of a duty is not actually a progressive way forward. (House of
Commons Environment Committee 1993b, Q700).
However, the idea to deliver energy efficiency through some sort of LCP approach using
utilities was endorsed by the House of Commons Environment Committee in a report on
energy efficiency which viewed ‘energy utilities and their regulators as key players in
107
Case I: Supplier Obligation in the UK
funding and implementing energy efficiency programmes’ and called for ‘a fundamental
shift in attitudes and approach on the part of the utilities and other power generators’
(House of Commons Environment Committee 1993a, para 5). The Committee also
travelled to the US to study LCP policies in various US states. NGOs such as Friends of
the Earth and Greenpeace supported the LCP approach and the idea of using utilities as
vehicles for the delivery of energy efficiency programmes. While British Gas clearly
resisted any move into the direction of LCP, the electric utilities reacted in a more
measured way with some utilities such as East Midland Electricity and Manweb
supporting LCP and carrying out pilot projects in that area. Other electricity utilities, such
as South Western Electricity and Eastern Electricity, were more concerned and reluctant to
embrace the concept of LCP (House of Commons Environment Committee 1993a).
The discussions bore fruit and according to the second year report on the Environment
White Paper, Government saw utilities in a good position to carry out energy efficiency
measures at the customer end (HM Government 1992).
E factor – the predecessor of the SO in gas
As a result of those discussions, in May 1991, Sir James McKinnon, the Director General
of OFGAS, announced a new gas price control formula to operate from April 1992 to
March 1997 with the purpose to encourage British Gas to compare the costs of supplying
gas with the option of investing in demand-reducing alternatives. This formula would
include an ‘E factor’ allowing gas suppliers to pass 1
% of the costs of energy efficiency
projects approved by the Director General through to gas customers. McKinnon expected
that around £50 million a year might be spent on energy efficiency measures (Owen 2006).
Money raised via the E factor was supposed to help fund the Energy Saving Trust (EST)
that was established by Government, British Gas and public electricity supply companies
108
Case I: Supplier Obligation in the UK
in 1992 to reduce home all energy use and the associated carbon emissions. In the early
days after its creation EST ran four energy efficiency projects such as a condensing boiler
scheme, a compact fluorescent lamps (CFLs) subsidy programme, residential combined
heat and power (CHP) schemes, and established a network of Local Energy Advice
Centres (Owen 1997). Although OFGAS established the E factor, the regulator made it
pretty clear from the beginning that it did not see wider environmental goals within its
remit and this ought to be dealt with by central government (House of Commons
Environment Committee 1993b). This would become an issue of great controversy later
on.
Summary
In the context of energy market liberalisation and quasi monopolies, policy makers and
stakeholders considered new policy instruments that used utilities as a vehicle for
promoting energy efficiency comparable to the LCP approach developed in the US. Once
the idea for an obligation on suppliers was on the table, feedback mechanisms reinforced
the implementation of the first schemes such as the E factor and EESoP 1. One feedback
mechanism was that as a market-based instrument in a context of liberalisation, obligations
on suppliers were characterised by a high degree of ideational fit. Market liberalisation
was seen as essential to promote economic efficiency and energy efficiency was very
much part of this idea as a means to ensure the efficient consumption of energy. It had
been expected that utilities would offer energy efficiency services as part of their portfolio
anyway once the market was fully liberalised. In the meantime, however, an obligation
was supposed to incentivise the utilities to deliver energy efficiency services (Interview
UK1, UK14). Therefore, obligations were fully in line with the dominant narrative at the
time. The circumstance that the SO (and the E factor) was an instrument independent of
109
Case I: Supplier Obligation in the UK
public expenditure comprised another feedback mechanism. Such schemes were seen as
having positive fiscal effects in the sense that they did not put additional strain on the
budget. This positive policy feedback mechanism prevailed and is probably one of the
main reasons why in times of deficit reduction programmes the SO remained unaffected.
4.2.2 Change of key personnel
Change of key personnel can trigger policy change in some instances and is a recognised
mechanism in the literature on policy change (e.g. Sabatier 1988). A good example of the
impact of changing key personnel affecting the dynamics of policy change can be found in
Great Britain during EESoP 1 and 2.
In the 1990s, when the SO was still run entirely by the regulator (including the targetsetting procedure), change of personnel at the top of the regulator OFGAS had a
significant effect on the development of the SO. The following part shows how a single
person, Claire Spottiswoode, who became director of OFGAS in 1993, successfully
blocked any attempts at implementing a SO in the gas sector and downgraded the E factor
significantly. In contrast, OFFER’s director was much more sympathetic to running energy
efficiency programmes and put in place the first SO in electricity in 1994. Only when a
new OFGAS director was appointed, was the way finally cleared for having a SO in gas as
well.
OFGAS’ role: obstruction of energy efficiency programmes
In general, McKinnon was ‘quite amenable to it [energy efficiency]’ (Interview UK ) and
it was likely that he might further advance the E factor over the coming years. However,
the story took a very different turn when McKinnon left office in 1993 and Claire
Spottiswoode took over. As a result of her appointment, in November 1993, there were
serious concerns that not enough money could be raised for EST making it impossible for
110
Case I: Supplier Obligation in the UK
the Agency to deliver the CO2 savings promised in the Climate Change Programme (DoE
1992). Claire Spottiswoode rejected raising money via the E factor for energy efficiency
measures administered by EST (ENDS 1994b). While the 1986 Gas Act did provide some
leverage for OFGAS to support energy efficiency measures, Spottiswoode did not feel
comfortable raising prices (OFGAS 1994). Spottiswoode took the view that ‘we should not
be interfering in the purity of the market with cross-subsidies of this kind’ (Interview
UK10). Once the market was fully liberalised, she argued, energy efficiency measures
would just be offered by the energy companies as part of a competitive market and
consumers were best placed to decide whether they wanted energy efficiency measures or
not. Spottiswoode expected that to happen in 1998 when the whole of the gas market had
been opened to competition. Only in the meantime would there be some very limited role
for energy efficiency measures as encouraged by the E factor (OFGAS 1994).
One of the interviewees described her as ‘just a classic, right-wing economist’ who
believed that if all those cost-effective energy savings opportunities existed, people would
do it anyway so why should one interfere with the free market (Interview UK4). Also,
Claire Spottiswoode argued that such decisions were within the realm of elected politicians
and could not be decided by the regulator (Interviews UK10, UK11). Furthermore, she
claimed that the E factor ‘was not introduced in order to deal with the
overnment’s CO2
remit’ but that it was ‘being hijacked into that’ (House of Commons Trade and Industry
Committee 1994, paragraph 93). She even accused her predecessor McKinnon of acting
illegally in authorising expenditure under the E factor (House of Commons Environment
Committee 1994), although she later withdrew those allegations following a legal review
by Jeremy Lever QC and Simmons and Simmons who deemed the E factor legal (Owen
1997).
111
Case I: Supplier Obligation in the UK
As a result, British Gas submitted a greatly scaled down package of EST projects to
OFGAS, but most of their proposals were subsequently rejected by the regulator (ENDS
1995). In the end less than £2 million compared to the £50 million announced by
McKinnon was spent by the time the E factor had ended in March 1997 (Owen 2006).
Following the same logic of argument, Spottiswoode also declined to put an obligation
similar to EESoP on gas suppliers. There was the feeling that without reforming the
primary legislation, i.e. the 1986 Gas Act, further energy efficiency programmes that
involved gas suppliers would be very unlikely (ENDS 1998).
OFFER’s role: measured support of energy efficiency programmes
The electricity regulator, OFFER, took a rather different view. OFFER issued a
consultation paper in 1991 on options of energy efficiency programmes including
Standards of Performance (OFFER 1991). However, the consultation paper remained
rather vague on specific instruments and also OFFER’s response to the consultation did
not contain any specific proposals (OFFER 1992). OFFER’s Director
eneral, Professor
Stephen Littlechild, suggested in 1992 the introduction of an energy efficiency scheme for
electricity similar to the E factor for gas (see above) (Eadie 1992). In November 1992,
OFFER announced that it planned to take energy efficiency considerations fully into
account during the 1994-1995 price control reviews (Woolf 1993). One interviewee said
that ‘there was the feeling that Steven Littlechild was playing catch-up’ with what was
happening in the gas sector at the time (Interview UK9). OFFER commissioned a study in
1992, which examined the potential for cost-effective utility-run energy efficiency
programmes. The study estimated that at least 6% of existing electricity use could be
realistically saved through efficiency improvements within ten years (LE Energy Ltd and
SRC International 1992). In 1993, OFFER circulated an unpublished discussion paper on
112
Case I: Supplier Obligation in the UK
the potential design of Energy Efficiency Standards of Performance with more detailed
proposals (OFFER 1993).
EST gave advice to OFFER on how EESoP should be designed. Part of this advice was a
study tour in the United States to learn more about Least-cost-Planning and Demand Side
Management in the US (Energy Saving Trust 1994). The level of EESoP proposed by EST
was followed by OFFER and all companies were allocated a target in line with EST’s
recommendation (OFFER 1994).
Eventually, in 1994, OFFER put energy efficiency standards of performance (EESoP) on
the Public Electricity Suppliers (PESs) (OFFER 1994). PESs were the 14 companies
created when the electricity market in the United Kingdom was privatised in 1990. This
was going to be the first SO and the scheme was administered by EST. EESoP 1 raised
about £100 million for energy efficiency projects, equivalent to £1 per customer per year
(Owen 1997). However, OFFER did not raise the amount of money expected by ministers
and OFFER’s Director
eneral had concerns that higher obligations ‘would raise issues
more appropriately dealt with through general fiscal policy’ (ENDS 1994a). OFFER made
this very clear from the beginning long before the scheme commenced in April 1994
(House of Commons Environment Committee 1993b). Also the second EESoP scheme,
which operated from 1998-2000, did not raise significantly more funds than EESoP 1.
While OFFER’s director had taken a more proactive approach to energy efficiency
programmes, Steven Littlechild made it very clear early on that he would only be willing
to do this at a discretionary level of not more than £1 per customer (Interview 4). Any
further increase in the costs of the SO to customers, he argued, would not be in line with
OFFER’s remit as an unelected (economic) regulator and therefore require government
intervention in some shape or form. So it was very clear at the time that if government
113
Case I: Supplier Obligation in the UK
wanted a SO at a much larger scale delivering higher reductions in all energy use and
carbon emissions, a rather substantial legislative change had to take place shifting the
target setting powers to central government.
Change of personnel and merger of regulators
The appointment of a new OFGAS director would eventually break the deadlock. Finally,
in September 1998, Claire Spottiswoode’s term as Director eneral of OF AS came to an
end and Callum McCarthy was appointed to be her successor. From the beginning,
McCarthy was much more sympathetic to having schemes such as the SO in gas
(Interviews UK3, UK4) and subsequently extended the SO to the gas sector
Prior to the extension to the gas sector an important change with regard to the regulator
took place: OFFER and OFGEM were merged into a single gas and electricity regulator
(OFGAS). While OFGEM was formally created by the Utilities Act 2000, OFGEM
already operated informally in 1999 with Callum McCarthy being joint director general of
OFGAS and OFFER from January 1999. Shortly after winning the elections in 1997, the
new Labour government announced its intention to merge the two regulators in March
1998 in a consultation document on utility regulation in general, arguing that it was
necessary to have a single energy regulator ensuring greater consistency due to the
converging gas and electricity markets with companies becoming dual fuel suppliers after
market liberalisation (DTI 1998b). This was confirmed later in the response to the
consultation in July 1998 (DTI 1998a) and proposals for legislation were finalised in 1999
(DTI 1999). It became law in 2000 as part of the Utility Act. In the meantime, Callum
McCarthy was joint director general of both regulators.
114
Case I: Supplier Obligation in the UK
Extension to gas
Callum McCarthy was much more cooperative with regard to energy efficiency schemes.
In 1999, EST urged Government to extend EESoP 2 as it was coming to an end in 2000. In
addition to getting an extension of the SO, EST hoped that gas suppliers would also be
covered (ENDS 1999a). The House of Commons Environmental Audit Committee took
EST’s concerns forward and recommended the extension of EESoP plus a separate scheme
for gas suppliers (House of Commons Environmental Audit Select Committee 1999).
Within just four days of publication of the Committee's report, the newly created OFGEM
issued proposals to extend the electricity EESoP scheme for two years from April 2000
and also proposed a new EESoP scheme for gas suppliers for the same period with a
savings target even higher than the electricity SO (ENDS 1999b; OFGEM 2000b). This
was a significant shift considering the gas regulator’s past opposition to an EESoP type
scheme for gas.
Summary
What this episode illustrates very clearly is that a single person’s objection to a policy
instrument, in this case the SO, can have a significant effect on whether this instrument is
adopted or not, an effect that is generally recognised in the literature on the theories of
policy change (Sabatier 1988; Sabatier and Jenkins-Smith 1993). The replacement of
Claire Spottiswoode with Callum McCarthy was definitely the main reason why the SO
was applied to the gas sector as well (Interview UK4). This process can also be understood
by applying the concept of friction: Spottiswoode was supposed to establish domestic
sector energy efficiency programmes in order to help with achieving the national carbon
reduction targets set by government (see section 4.2.5). However, she acted as a bottleneck
blocking most of the pressure put onto the regulator. As a result, policy inputs (such as
115
Case I: Supplier Obligation in the UK
demands to promote energy efficiency in order to reduce carbon emissions) were not
translated into policy outputs (in the form of an effective E factor or a SO for the gas
sector). Hence Spottiswoode’s objection was a source of friction leading to the
accumulation of pressure to reduce carbon emissions. EST frequently raised concerns that
without the promised funding from the E factor or a SO in the gas sector it would be
unable to generate the carbon emission reductions of the scale required. The fact that
shortly after his appointment McCarthy implemented the SO in the gas sector indicates
that at the time Spottiswoode was the primary source of friction rather than cognitive
limits or constrained institutional capacities as theorised by PET (Baumgartner 2011; Jones
and Baumgartner 2005, 2012). However, as the next section will show, there were also
significant institutional constraints, which caused substantial friction.
4.2.3 Transfer of target-setting powers
As discussed in chapter 2, one of the levers of policy change identified in the literature is
venue change. Venue change refers to situations when the institutional location of decision
authority changes (Baumgartner and Jones 1993). Such a venue change occurred in the UK
in 2000.
The debate outlined above shows that there were considerable conflicts about the
responsibilities and duties of the regulators in the context of energy efficiency and, more
generally, in environmental and social issues. Both regulators, OFGAS and OFFER, saw
themselves primarily as economic regulators and were not willing to ramp up the SO
significantly as they felt this was something government ought to be doing (Interview
UK , UK5, UK1 , UK11). At that stage an ‘impasse had been reached at which the
previous government had expected the regulator to do something and then the regulator
had not’ (Interview UK1 ). Without change in primary legislation, it looked like further
116
Case I: Supplier Obligation in the UK
measures would be extremely difficult to implement and this is exactly what happened in
the end when the new Labour government decided to shift the target-setting powers from
the regulator to the Secretary of State by primary legislation.
Proposals for utilities regulation reform
The new Labour government made clear from the beginning that they were going to take
action on utilities regulation and energy efficiency policy. Without a change in government
it would have been difficult to overcome the deadlock that had been reached in the 1990s
(Interview UK1, UK4).
In the 1998 review of the future of utility regulation, Government concluded that statutory
guidance from the government to the regulator needed to be issued to encourage the
regulator to take into account social and environmental considerations, and that the
regulator should be placed under a new secondary duty to have regard to that guidance.
However, government acknowledged that such guidance could not be used to implement
social or environmental measures (including energy efficiency measures) which have
‘significant financial implications for consumers or for the regulated companies’ (DTI
1998a). This is also reflected in the consultation on the 1998 UK Climate Change
Programme where it is argued that setting more ambitious targets for the SO ‘would fall
into the category of measures which should be implemented through new specific legal
provision rather than relying on guidance to the regulator’ (DETR 1998, para 66).
As a logical next step, in its proposals for legislation on utility regulation, government
announced in 1999 that it would put in place new legal provisions for government to set
the target of the SO (DTI 1999).
Utilities Bill
117
Case I: Supplier Obligation in the UK
In 2000, the proposals were passed through parliament as part of the 2000 Utilities Bill. As
expected, the Bill allowed the Secretary of State to issue guidance on social and
environmental matters to regulators. The first guidance of such kind was issued by DTI in
November 2002 but remained rather vague about home energy efficiency policy.
Succeeding guidance issued by DTI, later BERR then followed by DECC, did not include
stronger and more precise provisions and in 2010 the only reference to home energy
efficiency was that ‘the Authority should have regard to the fact that improved household
energy efficiency can also reduce costs to consumers and help to reduce fuel
poverty’(DECC 2010e, p. 10). In sum, the guidance did not have much impact on the SO
or in general (Interview UK1).
Furthermore, the primary duty of the 1986 Gas Act and the 1989 Electricity Act was
changed in the Utilities Act. While the primary duty previously was to promote
competition, the Utilities Act changed it to protect the interests of consumers including
future consumers, a change that some thought might be an important lever for more
ambitious environmental and social initiatives by the regulator, but this did not happen
(Interview UK1).
However, and more importantly for the development of the SO (Eyre and Staniaszek
2005), the Bill also gave the Secretary of State powers, by order, to impose energy savings
targets on gas and electricity utilities. As part of this, provisions were made in the 1986
Gas Act and the 1989 Electricity Act to impose an energy efficiency target on each gas
transporter and supplier, and each electricity distributor and supplier. This resolved some
of the conflicts over EESoP 1 and the E factor, which struggled to get the regulator’s
support due to conflicting duties and unclear responsibilities. The Utilities Bill directed the
responsibility for target setting to Government and obliged the regulator to implement the
118
Case I: Supplier Obligation in the UK
schemes that would be needed to reach the targets. OF EM’s role changed from being the
decision maker on the SO to becoming a ‘middleman’ having to put the scheme according
to government’s specifications on the energy companies (Interview UK6). This enabled
Government to significantly extend EESoP imposing much more ambitious targets on
energy companies.
Also, the Utilities Act was intended to ‘depersonalise’ the regulators by creating the
as
and Electricity Markets Authority (GEMA), essentially a governing board that replaced the
former Director General, in order to avoid some of the issues discussed in the previous
section (Lawrence 2000).
First SO after Utilities Bill
Shortly after the Utilities Bill had been passed, the government set the first SO target in the
Electricity and Gas (Energy Efficiency Obligations) Order 2001. Whereas the target of the
old EESoP 3 scheme was only 11 TWh (4.9 TWh Electricity & 6.1 TWh Gas) for a period
of two years (2000-2002), its successor, EEC 1, imposed a much higher target of 62 TWh
on energy suppliers (OFGEM 2005) for a period of three years (2002-2005), which
equates to a target four times as high as the EESoP 3 target.
The transfer of target-setting powers clearly was a key moment in the development of the
SO and without this institutional change it would be hard to imagine an SO at the level
where it is today. Asked whether the transfer of target-setting powers had been an
important moment in the history of the SO one interviewee said:
Oh crucial, we’d still be a pound a customer or one pound twenty a customer whatever it was.
We’d be stuck at maybe two pounds about now. And there was definitely a changing realisation
that this wasn’t just going away, this was a real problem. So yes. (Interview UK4)
119
Case I: Supplier Obligation in the UK
In a sense, although unintentional, the resistance of Claire Spottiswoode in the early 1990s
led in the end to a much stronger SO. Asked about whether there had been much resistance
from the different stakeholders to the shift of target setting powers, another interviewee
said:
No, my impression is not. Because the suspects that would have are precisely the ones who got
us into that position anyway, i.e. the rightwing let’s not interfere in the market people had been
the people saying we can’t have the regulator doing that because that’s politics, that’s putting
tax and politicians have got to do that. So when the government responded by saying politicians
will do it end of argument really. (Interview UK1)
However, representatives of the energy suppliers bemoaned that previously the ‘regulator
had a lot more freedom to tweak the rules and make sure they had the right outcome’
whereas since the government sets the target it has become much more ‘politicised’
(Interview UK1 ) and ‘high profile’ (Interview UK13) with ministers interfering with the
scheme more and more. While in the 1990s the SO was quite light touch and based on
gentlemen’s agreements, it appears to have come increasingly under the spotlight of
ministers and technical experts as the scheme moved along.
Nevertheless, the new institutional setting gave clarity to all the stakeholders involved in
this and put an end to the controversies around the remit of the regulator:
There were bits of the story that were about regulating an industry and there were other bits of
the story that were about energy efficiency, helping the poor, and by separating the decisions
they gave clarity. (Interview UK3)
In retrospect, the hesitance of the regulators proved an important driver for the institutional
change that took place, allocating the responsibilities in a manner that could easily be
understood by all stakeholders.
120
Case I: Supplier Obligation in the UK
Summary
The venue change that occurred in Britain resulted from a lengthy political debate about
the role of the energy regulators and their unwillingness to increase expenditure for energy
efficiency measures. In order to realise substantial carbon emission reductions in the
domestic sector modifying the institutional framework appeared to be the only option and
the accumulation of pressure eventually led to reform. Hence the venue change was a
deliberate means to achieve carbon reduction goals in the housing sector rather than an
unrelated coincidence. With regard to the conceptual framework of this thesis, the
following observations can be made. Due to the inability of existing institutions (in this
case OFGAS and OFFER) to raise the energy savings target, the pressure to generate the
carbon emission reductions elsewhere accumulated. Directing the target-setting powers to
the central government instead of the regulator was seen as the only way of dealing with
the pressure to achieve carbon emission reductions in the housing stock. Analysed through
the lens of friction, the institutional constraints (unwillingness and inability of OFGAS and
OFFER to increase efforts beyond a low level) limited the system’s ability to translate
policy inputs (pressure to cut domestic carbon emissions) into policy outputs (effective
SO).
4.2.4 Energy prices
Energy price shocks are often cited as an example for a crisis that leads to a modification
of existing policy instruments (Campbell 2004). The oil crisis in 1973 is a good example
of policy change induced by crisis in the form of non-linear price signals. Whereas before
energy efficiency was of very limited political concern, this changed almost overnight
(Eyre 1997). As expected, energy prices played an important role for the development of
the SO. Rising prices are clearly linked to fuel poverty, another policy driver assessed in
121
Case I: Supplier Obligation in the UK
section 4.2.6, but rising prices constituted a policy driver due to concerns about their effect
on all consumers beyond low-income households (Interview UK14).
In the 1980s and 1990s, energy prices were low compared to the 2000s with the general
trend of falling prices since the early 1980s (Figure 6). However, from 2004 onwards,
domestic energy prices of all fuels increased dramatically. Gas prices in real terms almost
doubled in just a decade; 3 prices for coal and smokeless fuels increased by 65% and
electricity prices by 45%.
Figure 6: Fuel price index numbers relative to the GDP deflator (2005=100)
170
Coal and smokeless fuels
Gas
Index 2005=100
150
Electricity
130
110
90
70
1970
1975
1980
1985
1990
1995
2000
2005
2010
Source: based on DECC (2011f)
This trend generally pushed energy policy up the political agenda, as one interviewee
pointed out:
3
Used fuel price index relative to the GDP deflator.
122
Case I: Supplier Obligation in the UK
We had relatively abundant domestic energy in the UK, relatively benign energy prices, and, I
mean compared to today, completely different world and energy policy was just not seen as that
important. (Interview UK2)
While energy prices are a direct driver of fuel poverty (see section 4.2.6), reducing the
costs of energy services in general was a separate key objective of the SO from the
beginning. References in the various consultation documents confirm that with earlier
schemes referring to the efficient consumption of energy based on the duties of the
regulator laid down in the 1986 Gas Act and the 1989 Electricity Act (OFFER 1994,
1998a). EEC 1’s objectives included helping consumers to lower their fuel costs or
increase their comfort as one of three main objectives alongside reducing fuel poverty and
carbon emissions (DEFRA 2001). Similarly, the EEC 2 and CERT consultation documents
name lowering the costs of all energy use in households as a key objective of the SO
(2004a, 2007a), separately from fuel poverty. It therefore comes as no surprise that rising
energy prices had an impact on the development of the SO.
Sharply rising energy prices in 2008
Prices increased rapidly during the winter of 2006/2007 and 2007/2008. The rising energy
prices particularly affected the SO in 2008, when prices increased more quickly than any
increase since the 1973 oil crisis. However, in 2006/2007 no changes took place even
though there were some discussions about high energy prices (e.g. Hussain 2006) although
with a significantly lower intensity than in 2008.
As shown in Figure 7, residential gas prices in 2008 increased almost by 50% in real terms
in just four quarters. This was mainly a result of rising wholesale gas prices in continental
Europe, where gas prices are contractually linked to oil prices. Similarly, electricity prices
went up by almost 30%, also mainly due to rising wholesale gas prices.
123
Case I: Supplier Obligation in the UK
Figure 7: Fuel price indices in the domestic sector in real terms Q1 2000 to Q1 2010
180
Coal and smokeless fuels
Index 2005=100
160
Gas
Electricity
140
120
100
80
Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
Source: based on DECC (2010d)
The developments in the energy markets triggered a controversial discussion about energy
companies and their pricing as well as their profits, which in the end led to a change of the
SO.
Discussions about a windfall tax in 2008
In the context of sharply increasing energy bills, there were calls in early 2008 for a
windfall tax on energy suppliers. A heated discussion started after OFGEM reported to the
Treasury and the Committee on Business and Enterprise that energy companies made £9
billion profit from EU ETS permits which were issued for free. In January 2008, OFGEM
suggested that this windfall could be used to fund help for the fuel poor (OFGEM 2008b).
Just a few weeks after the OFGEM proposal, Government held a meeting with the heads of
major energy suppliers and told them that they might face a levy on their profits to help the
124
Case I: Supplier Obligation in the UK
poor (Oliver 2008). As expected, the proposals were not met with great enthusiasm by the
energy suppliers. In February 2008, OFGEM announced a two-year investigation into the
price rises and the competitiveness of the markets in electricity and gas for households and
small businesses (OFGEM 2008a). In early March 2008, the Local Government
Association, a cross-party organisation representing councils in England and Wales,
proposed to require utilities to pay £500m annually for home energy efficiency measures
(LGA 2008). There were expectations that the 2008 Budget might introduce a windfall tax
but this turned out not to be the case (Grice and Russel 2008).
However, discussions about a windfall tax continued and in July 2008 the Committee on
Business and Enterprise argued that ‘there is a compelling rationale for at least a modest,
one-off top-slicing of these gains to help fund action to reduce the energy bills of
vulnerable families in the long term’ (House of Commons Business and Enterprise
Committee 2008, p. 47). Increasing profits made by energy suppliers such as Centrica after
raising domestic fuel prices by 35% led to a renewal of calls for a windfall tax in August
2008. There was, however, no agreement in Government on the matter –Chancellor
Alistair Darling and John Hutton, the Business Secretary, opposed a windfall tax on the
basis that energy companies needed extra funds to expand low-carbon energy sources
(Kennedy and Pagnamenta 2008; Webster and Elliott 2008). After EDF pulled out of a
deal which included buying British Energy and probably constructing new nuclear power
plants, there were signs that a windfall tax was off the agenda (Wighton 2008).
However, more than 70 Labour MPs signed a petition that called for a windfall tax and
hoped to make the issue one of the dominating themes of the party’s annual conference in
September. Among them was Geoffrey Robinson, a key ally to Gordon Brown who
devised the windfall tax on energy companies in 1997 (Stratton 2008). Shaun Woodward,
125
Case I: Supplier Obligation in the UK
the Secretary of State for Northern Ireland, even threatened to resign if there was no
windfall tax coming (Webster and Elliott 2008). A windfall tax was also supported by the
unions. As expected, energy companies opposed any attempts to introduce a windfall tax
and argued that they would need to raise prices even further in case of a tax due to the need
to build new power plants (Jameson 2008). In early August 2008, Government examined
different alternatives to a windfall tax. One of them was to oblige the energy companies to
spend the bulk of the money under CERT in the first two years and also to increase the
share spent on helping the fuel poor. Raising the proportion of auctioned EU ETS permits
and a carbon levy was also considered (Wintour 2008). Government eventually designed a
£1bn package of new funding and measures to tackle fuel poverty and threatened the
energy companies with a windfall tax in case they did not sign up for the package (Webb
2008).
Substitution for windfall tax: CERT uplift
On 11 September 2008, Government finally revealed the package they negotiated with the
energy companies – the Home Energy Saving Programme. The package included an
increase in the existing CERT target by 20% with a new target of 185m lifetime tonnes of
CO2 for the period April 2008 - March 2011. That implied additional expenditure by the
energy suppliers of an estimated £560m (HM Government 2008). Alongside this,
Government introduced the new Community Energy Savings Programme (CESP),
increased funding for the Warm Front programme, and a range of specific fuel poverty
measures such as tripling the Cold Weather Payments 4 and increasing the Winter Fuel
4
Cold Weather Payments are automatically paid to people on certain benefits when the temperature
in the area they live has dropped or is forecast to drop to zero or below for 7 consecutive days.
126
Case I: Supplier Obligation in the UK
Payments.5 overnment argued that increasing CERT and introducing CESP ‘is the better,
simpler way to meet our objectives than a windfall tax’ and that ‘by choosing this route the
Government can more swiftly help families cut fuel bills now and in the medium term;
help secure the long-term investment in new low-carbon energy infrastructure this country
requires; and help keep prices down’ (HM Government 2008, p. 2). The uplift of CERT
was subsequently put into law in July 2009 in the Electricity and Gas (Carbon Emissions
Reduction) (Amendment) Order 2009.
While a DECC official claimed that the decision against a windfall tax and the uplift of
CERT were not directly related and that the CERT uplift was ‘a very separate decision’
(Interview UK6), it seems unlikely that this was the case as confirmed by various
interviewees (Interview UK1, UK6, UK8, UK10, UK12). One interviewee made this very
clear:
[..] definitely, the package of not having a windfall tax included increasing CERT, bringing
CESP in, all these sort of things that were announced at the time that was dressed up as ‘we’re
doing this instead of a windfall tax’. (Interview UK12)
Favourable outcome for energy companies
Not surprisingly, the energy companies made it very clear to the government that they
were opposed to a windfall tax in negotiations with government:
I think there were quite a lot of conversations behind closed doors, if you know what I mean.
You had energy companies going to talk to officials, ministers being spoken to, quite strong
lobbying. I think the mood music played from what actually happened was very much, along the
5
Winter Fuel Payment is an annual tax-free payment made to people older than 60 years to help
towards their winter heating costs.
127
Case I: Supplier Obligation in the UK
lines, the energy companies were very clear, they did not want a windfall tax on their profits.
(Interview UK5)
Apparently, the energy companies did well in conveying their arguments against a windfall
tax:
Let’s just say as an energy supplier I’d be really proud of my negotiation skills on that. To
manage to go from a windfall tax to what was a bit of an extension of CERT was impressive
negotiations. (Interview UK8)
It seemed that the concerns raised by the energy companies regarding the risks of a
windfall tax to long-term investment in the energy infrastructure convinced the
Government.
Proponents of a windfall tax criticised the Government that the announced measures would
not go far enough. However, there was also support for the measures, for example from the
Association for the Conservation of Energy(ACE) (ENDS 2008), a lobbying group
representing the interests of the energy efficiency industry. The Association of Electricity
Producers announced that the costs of the package would be passed on to consumers by
energy companies which the Prime Minister promised would not happen (Naughton 2008),
although the architecture of CERT clearly contradicts that statement in that cost passthrough is not regulated and down to the discretion of the energy suppliers.
The outcome of the debate, the uplift of CERT and introduction of CESP compares quite
favourably to a windfall tax: In case of the windfall tax the government would have taken
away some of the profits of the energy companies whereas any rise of expenditure for
CERT could be passed through to consumers. Also, because the energy companies did not
have to pay for the CERT uplift but the package was sold as a burden on energy suppliers,
128
Case I: Supplier Obligation in the UK
they could claim that they helped households in fuel poverty in times of high energy prices
and use this as part of their PR (Interview UK6).
Intervention of the Prime Minister
One point worth noting is that the decision to increase the CERT target was related to an
intervention of the Prime Minister. Gordon Brown was directly involved in the
negotiations with the energy companies and the involvement of No. 10 clearly had an
impact on the modifications that CERT was undergoing.
One interviewee commented on the proposals of the Brown government as following:
I remember the frantic response of the Brown government in 2008 when the oil price hit $147
and energy efficiency was suddenly on top of the government’s agenda. (Interview UK10)
Apparently, DEFRA were rather slow to respond to the political pressure and, as a result of
this, No. 10 took over the process (Interview UK10). Gordon Brown was under a lot of
pressure with the Labour Party conference coming up at the end of September 2008
(Interview UK12). Government departments were asked about potential policies to address
the peaking energy prices and in particular domestic fuel bills. DEFRA recommended
using the existing CERT mechanism because ‘it had clear legs, it was something that could
be delivered quickly’ (Interview UK6).
So there were at least two reasons, why the CERT uplift had been chosen as the main
response to the mounting political pressure: First, the energy companies agreed to the
CERT uplift as opposed to a windfall tax. Second, the SO had been in place for 14 years at
the time and the government knew it could be expanded easily.
When Gordon Brown announced the Home Energy Saving Programme, he apparently
made an error. In his statement he said that over the next three years six million homes
129
Case I: Supplier Obligation in the UK
would be insulated at a rate of two million a year. However, the PM’s statement was based
on a draft by DEFRA which said that the Home Energy Saving Programme would deliver
energy savings equivalent to what would be produced by insulating 6 million homes, a
very different target definition because it does not prescribe the number of homes to be
insulated, but sets a carbon reduction target. Nevertheless, once the statement was made
Gordon Brown insisted on the target he announced (Interview UK7, UK10, UK11, UK12).
After the Prime Minister’s statement, the government set up a special task group to make
sure it was delivered. The task group comprised the insulation industry and the energy
suppliers and was called ‘Action on Insulation’ (Interview UK7).
Summary
Attempts to introduce a windfall tax on energy companies’ profits in
8 were
unsuccessful due to resistance from within government and the companies themselves.
Alternatively, the government decided to raise the CERT target and to introduce CESP.
Considering that CERT had just started in April 2008, the announcement of the 20% uplift
in September 2008 was certainly unusual and unprecedented. Due to considerable friction
the proposals for a windfall tax did not bear fruit, but at the same time the pressure to
respond politically to rising energy prices and increasing profits of energy companies
accumulated. Eventually a venue change occurred and Gordon Brown dealt with the issue
directly. Without the Prime Minister’s intervention the
% uplift of CERT would have
been rather unlikely, given that CERT was running for just about five months and a change
to the target half way within a SO scheme had never happened before. The intervention of
macropolitics in subsystem politics is one of the main causal mechanisms leading to policy
change hypothesised by PET (Baumgartner and Jones 1993; True 2000; True et al. 2007a;
Walgrave and Varone 2008). Macropolitics can overcome inertia inherent in policy
130
Case I: Supplier Obligation in the UK
subsystems as a result of friction because ‘at the macropolitical level, old and new political
interests can be rebalanced and major redirections of policy can be launched’ (True 2000,
p. 12). In this case the involvement of the Prime Minister himself led to significant and
rapid changes in the subsystem of domestic energy efficiency policy.
4.2.5 Climate change
Climate change policy became the most important driver of the SO over time. However,
the significance of it gradually increased over the years.
First references to climate change
As outlined above, the origins of the SO have a lot to do with discussions around
economic efficiency/Least Cost Planning and less with climate policy. One interviewee
made this very clear:
‘When we thought of energy efficiency we were still very much in terms of improving the
efficiency of energy use and reducing energy intensity as opposed to saving tons of carbon.’
(Interview UK10)
However, ‘those sort of rather arcane regulatory economics issues became less important
and it became much more about climate’ (Interview UK1). Climate change climbed the
political agenda in the UK over the last two decades and demarcated a distinct policy area
after Margaret Thatcher mentioned climate change as a substantial challenge to mankind in
her 1988 landmark speech to the Royal Society (Pearce 2006). Thatcher famously
suggested that ‘we have unwittingly begun a massive experiment with the system of this
planet itself’ (Thatcher 1988). Her speech was seen as a sharp turning point in her
government’s approach towards environmental policy (Interview UK3, UK1 ) and drew
the public’s attention to the issue of climate change (Anderson 1991).
changing focus on climate change was also restated in the 199
131
overnment’s
White Paper ‘This
Case I: Supplier Obligation in the UK
Common Inheritance: Britain's Environmental Strategy’ (HM Government 1990) which
committed the UK to stabilise its carbon emissions at 1990 levels if other countries would
join in.
The increasing importance of climate change was also reflected in the way the SO was
perceived. For example, as part of the first report under Article 12 of the Framework
Convention on Climate Change, which was ratified in 1993 (HM Government 1994), the
Energy Efficiency Office had to calculate the amount of carbon emissions that would be
saved by various policy instruments including the SO (Interview UK10). Together with the
E factor, EESoP 1 was supposed to raise money for the EST that was established by
Government, British Gas and public electricity supply companies in 1992 to reduce home
all energy use and the associated carbon emissions. The EST played a key role in the
overnment’s climate policy strategy for the domestic sector as outlined in the 1994 UK
Climate Change Programme (HM Government 1994), and the government wanted EESoP
1 to play a leading role in reducing domestic sector carbon emissions (Royal Commission
on Environmental Pollution 2000). This was driven by the recognition that energy
efficiency offered a lot of ‘low hanging fruits’ in the form of low-cost carbon reduction
options (Interview UK2).
While the issue of climate change was around in the early 1990s, it was not a major driver
of the SO (Interview UK1, UK3, UK7, UK11, UK12, UK13, UK14). As one interviewee
put it: ‘all of that stuff that now dominates government thinking we were just beginning to
introduce’ (Interview UK1 ). Climate change ‘was still very much a subsidiary business,
the main business was about a competitive free market in energy, and cheap energy and
[…] let the market sort it out’ (Interview UK3). The comparatively low significance of
132
Case I: Supplier Obligation in the UK
climate change in the 199 s was also reflected in the energy companies’ marketing of their
energy efficiency programmes:
‘All the marketing at the time was all about making customers more comfortable and saving
money. I think we downplayed the environmental benefits because quite a lot of research
showed that customers weren’t interested at that time.’ (Interview UK13)
The low importance of climate change is also reflected in the consultation paper by
OFFER on energy efficiency (OFFER 1991) and the document setting out the first EESoP
scheme (OFFER 1994): There is not a single reference to climate change or carbon
emissions. Four years later, only very few references to carbon emissions can be found in
the EESoP 2 consultation document, mainly in the context of cost and lifetime savings
(OFFER 1998b). The final EESoP 2 document does not make a single reference to carbon
emissions or climate change (OFFER 1998a).
Climate change climbing up the agenda
However, climate change stayed on the government’s and the public’s agenda. One
indicator that can be used to track back the role that a particular issue played in discussions
at the time is media coverage. Media coverage is often used to analyse political pressures
as they ‘influence policymaking through the highlighting of issues - designating some
important and others less important’ (Jones and Baumgartner 2005, p. 211).
An analysis of the number of newspaper articles in The Guardian and The Observer, The
Times and The Sunday Times, and The Independent and The Sunday Independent clearly
shows that climate change was reported on increasingly over the 1990s, with peaks after
Thatcher’s 1988 speech and the 199 Rio Summit as well as the 1997 Kyoto Protocol
(Figure 8).
133
Case I: Supplier Obligation in the UK
Figure 8: Newspaper coverage of climate change in the UK
Number of newspaper
articles on climate change
2500
2000
1500
1000
500
0
1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Source: based on Boykoff and Rajan (2007)
This trend continued until 2006 when it started to slow down and media coverage declined
again, although there was another peak in 2009 due to the Copenhagen climate summit and
the Climatic Research Unit email controversy (also known as ‘Climategate’) (Boykoff
2011; Painter 2011).
While there was increasing media coverage, important policy changes took place as well.
Particularly the Kyoto Protocol on climate change and the agreement of the newly elected
Labour government to a 12.5% reduction in greenhouse gas emissions by 2008-2012 based
on 199 levels put climate change firmly on the government’s agenda. This was followed
by a new Climate Change Programme in 2000 committing the UK to a 20% reduction of
carbon emissions by 2010 based on 1990 levels, a target that had featured previously in the
Labour Party’s 1997 election manifesto (Labour Party 1997). The draft of the 2000
Climate Change Programme explicitly mentions the SO as one of the six key UK policy
measures to save carbon (DETR 2000). However, at the time the targets were
comparatively humble and nothing of the scale seen later in the 2000s. The idea of having
134
Case I: Supplier Obligation in the UK
more long-term targets was not around at that time (Interview UK1) and it took some time
until climate change targets became much more ambitious.
Nevertheless, for the first time since the commencement of the SO, the EESoP 3
consultation paper explicitly mentioned climate change and referred to the Kyoto target
and the 20% reduction target set out in the Labour election manifesto (OFGEM 1999).
First ambitious emission reduction targets
The starting point to this is probably a report published by the Royal Commission on
Environmental Pollution in 2000 on energy and climate change which recommended a
60% reduction in carbon emissions by 2050 based on 1990 levels (Royal Commission on
Environmental Pollution 2000). The figures of the report were based on a June 1996
decision of the EU Council of Ministers as outlined in the EU Community Strategy on
Climate Change to limit emissions to such an extent that atmospheric CO2 concentrations
of 550 ppm would not be exceeded. This figure, in turn, goes back to the 1995 IPCC
Second Assessment Report (IPCC 1995) which argued that atmospheric CO2
concentrations needed to be kept below 550 ppm in order to limit the global average
temperature rise to around 2 degree Celsius above the preindustrial level.
As one of the first areas with significant potential to help to achieve such a reduction the
report highlighted the potential to lower the level of all energy consumption via efficiency
measures. The report argued that the ‘scope for improvements in buildings of all kinds, but
especially housing, is particularly large’ (Royal Commission on Environmental Pollution
2000, p. 4). While the report acknowledged that the SO could play a critical role in
reducing households’ all energy use, it also pointed out that existing policies were falling
short of achieving a significant enough reduction of all energy use and that further
measures were needed.
135
Case I: Supplier Obligation in the UK
The growing importance of climate change as a political driver was also reflected in a
modification of the metric: Under EESoP 1-3 the target was simply defined in kWh
savings to be achieved (although it was an indicative target, the actual target was set in £
per customer per year). EEC 1 introduced a target with fuel weighted kWh, i.e. it depended
on the carbon intensity of the fuel saved how much it would count towards the target. So
effectively, EEC 1 introduced a carbon target (Interview UK1, UK13).
When DEFRA presented its proposals for EEC 1, climate change was mentioned as one of
the main reasons why EEC 1 would be implemented. According to the consultation
document, one of EEC 1’s three objectives was to make a significant contribution to the
UK’s carbon targets (both Kyoto and the 20% target) alongside saving households money
and helping with the alleviation of fuel poverty (DEFRA 2001).
Energy efficiency and its potential contribution to reduce carbon emissions in the UK also
played a major part in an energy review carried out by the Cabinet Office’s Performance
and Innovation Unit (PIU). In the chapter on reducing carbon emissions the report argued
that the SO (at the time EEC 1) was ‘expected to deliver the largest carbon savings in the
domestic sector’ (PIU 2002, p. 113). Therefore, the review argued, DEFRA and DTI
should make an early commitment to extending EEC from 2005-2010 keeping it at least at
the level of EEC 1.
The above mentioned report by the Royal Commission on Environmental Pollution
suggested a reduction of carbon emissions of 60% by 2050 based on 1990 levels, a figure
that featured as an indicative target in the 2003 Energy White Paper. The White Paper
placed energy efficiency firmly on the political agenda - it was presented as the most
important measure in the short and medium term in order to achieve all four of the
objectives of the Paper (tackling climate change, security of supply, economic
136
Case I: Supplier Obligation in the UK
competitiveness and social equity) (Eyre and Staniaszek 2005). Because of its emphasis on
climate change and energy efficiency, the White Paper had been a turning point in UK
energy policy (Eyre 2008).
In line with the PIU review, the White Paper stressed that in order to achieve that sort of
scale of reduction the SO ‘will have a major role to play in homes’ (DTI 2003, p. 34). The
White Paper also announced that that the government would consult on an extension of the
SO from 2005 to at least 2008 with a target twice its level of activity under EEC 1. While
there are some references in the White Paper to fuel poverty when the SO is mentioned, it
clearly is framed as a policy instrument that has the primary objective of reducing carbon
emissions.
Effect of first carbon targets on SO
There is evidence that the climate targets directly impacted on the design of the SO. Prior
to the commencement of EEC 1, the government department responsible for setting up the
SO was assigned the task of linking the expected carbon savings from the SO to the overall
carbon reduction targets:
‘We were doing a lot of top-down analysis linking to the bigger picture White Paper and
Climate Programme targets where we were sort of saying ‘how many millions of tons of savings
do we need to deliver from this sector by 2010?’, initially to meet the Kyoto target and the UK
pledge to do a 20% CO2 reduction by 2010, and we portioned that out between the different
sectors. Then you have to say what contribution does the cumulative savings of all these
schemes have to make by 2010.‘ (Interview UK2)
From EEC 1 onwards, climate change appears to be the strongest driver and the
government targets put pressure on the departments to deliver a substantial contribution to
the carbon targets via the SO (Interview UK2, UK4, UK5).
137
Case I: Supplier Obligation in the UK
While the EEC 1 consultation document referred to climate change as one of three
objectives, the EEC
consultation document stresses that ‘the primary aim of EEC
is to
make a significant contribution to the UK’s legally binding target under the Kyoto protocol
[…] and its domestic goal to cut emissions of carbon dioxide by
% below 199 levels by
1 ’ (DEFRA 2004a, p. 5).
Another milestone was the Stern Review on the Economics of Climate Change
commissioned by the Treasury in 2005 and published in 2006. The report received a lot of
attention both nationally and internationally. Although it is primarily about the economics
of climate change, there are some references to energy efficiency. The review supported
the view of the International Energy Agency (IEA) that ‘by
5 , energy efficiency has the
potential to be the biggest single source of emissions savings in the energy sector’ (Stern
2007, p. 13).
The SO remained a key element in the government’s climate change strategy and featured
among the top three additional measures to save carbon across all sectors in the 2006
Climate Change Programme (HM Government 2006). This was confirmed in the 2007
Energy White Paper. Government expected that in 2020 the SO would deliver up to 17%
of all carbon reduction measures planned at the time, making it the second most significant
measure after the EU emissions trading scheme. The White Paper highlights the
government’s commitment to continue delivering carbon savings in the household sector
via the SO until at least 2020 (DTI 2007).
Changing metric of the SO
In order to align the SO with the wider climate policy landscape, the metric of the SO
changed from TWh to carbon emissions when CERT commenced in 2008 (Interview
UK6). The Climate Change and Sustainable Energy Act 2006 changed the provisions
138
Case I: Supplier Obligation in the UK
previously made through the 2000 Utilities Act in the 1986 Gas Act and the 1989
Electricity Act, which enabled the Secretary of State in the past to set an energy efficiency
obligation on each gas transporter and supplier, and each electricity distributor and
supplier. The altered sections gave powers to the government to set the obligations in the
form of a carbon emissions reduction target.
Another reason for the new metric was that government wanted to include additional
measures such as micro-generation and behavioural measures in the SO, although very
limited activity on CHP and solar thermal had already been carried out under previous SO
schemes (see section on mix of measures above). The official reasoning behind the
decision to open the SO to additional measures was as follows:
The Government wishes to encourage a holistic approach to carbon abatement in the household
sector, encouraging customers to take advantage of the full benefits of combining both energy
efficiency and microgeneration to reduce their carbon emissions. We also want to raise
awareness among household consumers of the impacts of their energy use and encourage them
to change their behaviour and attitudes to energy in the home. As part of this approach we wish
to provide suppliers with even more flexibility in the measures they can adopt to meet their
targets. (DEFRA 2006a, p. 11)
Although CERT focused explicitly on carbon emissions whereas EEC 1 and 2 defined the
savings target in terms of TWh, the change was mainly semantic. EEC used fuel-weighted
savings targets based on the carbon content of the various fuels, i.e. it was essentially a
carbon emissions saving target (Interview UK1).
DEFRA consulted on the additional measures and the change of metric shortly after the
Climate Change and Sustainable Energy Bill passed the last stage in the House of
Commons (DEFRA 2006a; ENDS 2006a, 2006c). Later, DEFRA published a draft list of
the additional measures that suppliers could use to reach their target (ENDS 2006b).
139
Case I: Supplier Obligation in the UK
In the consultation document on CERT, DEFRA made it very clear that the SO’s main
purpose was to contribute to the UK’s climate change policy:
‘[T]he primary aim of the CERT is to make a significant contribution to the UK’s legally
binding target under the Kyoto protocol to cut greenhouse gas emissions by 12.5% below 1990
levels by 2008–2012 and its domestic goal to cut emissions of carbon dioxide by 20% below
1990 levels by 2010.’ (DEFRA 2007a, p. 12)
Climate Change Act
In 2006, the UK became the first country in the world to commit itself to a legally binding
carbon reduction target of 60% in the Climate Change Bill. This followed a period of
pressure from a number of NGOs including the Big Ask Campaign of Friends of the Earth
which resulted in more than 200,000 people asking their MP to sign up to a binding target
of 8 %. Also the WWF’s
et on Board campaign supported an 8 % reduction target. The
Bill was drafted by Friends of the Earth and brought before Parliament on 7 April 2005 by
three MPs from the three major parties, namely Michael Meacher (Labour), John Gummer
(Conservatives), and Norman Baker (Liberal Democrats). Because parliament was
dissolved prior to the 2005 election, the bill was unable to make progress. However,
shortly after the election an Early Day Motion was signed by more than 400 MPs
demanding a climate change bill. The Prime Minister announced a climate change bill in
the
6 Queen’s Speech.
While the initial target was only a 60% reduction, the target was raised following a
recommendation by the Climate Change Committee, which was set up as part of the bill to
advise government on its climate change policy. After several debates on the bill, it finally
became law in November 2008.
140
Case I: Supplier Obligation in the UK
Part of the bill was the extension of the primary legislation via amendments in the 1986
Gas Act and 1989 Electricity Act allowing the Secretary of State to put carbon emissions
reduction targets also on energy generators. Such changes were required to establish the
Community Energy Savings Programme (CESP), an area based scheme targeting energy
generators as well as suppliers. The Climate Change Act did not make provisions that
affected the SO until 2010, but it gives government more flexibility to modify the SO in
the future.
Summary
Climate change clearly has been one of the main drivers for the development of the SO. Its
importance gradually increased over a long period of time. While the earlier SO schemes
made very little (if any) references to climate change, later schemes explicitly mention the
contribution of the SO to reducing carbon emissions as one of the key objectives of the
scheme. From EEC 2 onwards, the principal aim of the SO according to government
documents was to reduce carbon emissions in the household sector. Behind the growing
importance of climate change objectives for the SO lay a number of factors, including the
steadily increasing media coverage of climate change and the development of legally
binding carbon reduction targets resulting in the Climate Change Act in 2008. The
increasing pressure on the SO did not materialise until the target setting powers were
transferred from the regulator to the central government due to resistance from the
regulator to upscale the SO (see sections 4.2.2 and 4.2.3). Hence there was substantial
friction as hypothesised by PET, in particular in the 1990s. Following the 1997 Kyoto
target, high profile reports, such as the landmark publication by the Royal Commission on
Environmental Pollution in 2000, pointed out that current policies did not deliver
significant enough carbon savings, an indicator that the policy inputs (pressure to act on
141
Case I: Supplier Obligation in the UK
climate change) were not translated into policy outputs (policy instruments focusing on
carbon emission reduction). While the targets increased in each period after EESoP 3, the
government did not feel comfortable raising the targets more quickly and to the extent
required in order to reach the national climate targets. The hesitation was due to a lack of
experience with the SO and restricted administrative capability at the time (see section
4.2.8), another source of friction. Still, the 80% carbon reduction target adopted put
pressure on policy makers to expand the SO and eventually the SO became the second
most important climate policy after the EU ETS.
4.2.6 Fuel poverty
One of the drivers of the SO was fuel poverty. While this chapter cannot provide a
comprehensive account of the history of fuel poverty policy in the UK, its impact on the
SO is analysed and where appropriate the wider historical background is provided. For a
detailed analysis of fuel poverty policy in the UK please refer to Boardman (1991a, 2010).
In Britain fuel poverty is an important driver of energy policy and became a distinct issue
of public concern following the oil crisis in 1973-1974. The recognition of regressive
impacts of rising energy prices led to a distinction between fuel poverty and general
poverty (Bradshaw 1983). Boardman argued that ‘fuel poverty is different’ (Boardman
1991b, p. 30) from income poverty because of ‘the crucial role of housing stocks - the
house, heating system and other energy-using equipment’ (Boardman 1991a, p. 221).
Fuel poverty in the UK is defined as the need to spend more than 10% of household
income on all energy use in order to maintain a satisfactory heating regime and other
energy services (DECC 2011a). The original definition of fuel poverty goes back to
Boardman (1991a). According to DECC (2011a), the level of fuel poverty depends on the
interaction of three factors: income, fuel prices paid, and fuel requirement (which is
142
Case I: Supplier Obligation in the UK
affected specifically by the fuels used, the dwelling and energy efficiency). Hence fuel
poverty goes up if energy prices rise, although some of this can be offset by rising incomes
and increased energy efficiency. Note that at the time of writing the official definition of
fuel poverty used has been under review by John Hills (2011, 2012) and is likely to change
in the future.
Fuel poverty in the early days of the SO
The Conservative government did not recognise fuel poverty as a problem and avoided
using the phrase in official documents (Boardman 2010). Its focus was very much on
privatisation of the energy industry assuming that the benefits of this process would
eventually reach all customers in the form of cheaper energy prices (Sharratt et al. 2007).
This is also reflected in the regulator’s publications on energy policy at the time. Not
surprisingly, the consultation document on energy efficiency published by OFFER in 1991
does not refer in any way to fuel poverty (OFFER 1991). There is also no reference to fuel
poverty either in the consultation or the proposals on EESoP 1 (1993, 1994). However,
part of the criteria that the energy suppliers were required to consider when selecting
projects under EESoP 1 were related to fuel poverty. Suppliers were obliged to take into
account ‘the interests of Consumers and in particular the interests of those Consumers who
are elderly or disabled, who are in rural areas or who may have difficulty in meeting their
obligation to pay for electricity’ (OFFER 1994, p. 20). Those criteria are based on the
duties of the regulator laid out in the 1986 Gas Act and 1989 Electricity Act (see section
5.1.2). EESoP 1 was put in place with the expectation that around 35% of energy savings
would accrue to low income customers, although in practice about 6 % of the suppliers’
expenditure had been spent on this group (OFFER 1998a).
143
Case I: Supplier Obligation in the UK
Nevertheless, in the early 1990s the SO was not oriented towards fuel poverty very much
and there was a presumption that, if at all, then government should fund fuel poverty
programmes such as the Home Energy Efficiency Scheme, which later became Warm
Front (Interview UK11). Fuel poverty was seen as an unacceptable concept according to a
former DETR official:
[…] at the time we weren’t really allowed to talk about fuel poverty, that wasn’t an acceptable
issue to talk about. (Interview UK3)
However, while fuel poverty clearly did not play a significant role in the 1990s (both for
energy policy in general and the SO), in the 2000s a lot of debates took place around fuel
poverty and its eradication.
Change of government and more emphasis on fuel poverty
While the Conservative government more or less ignored fuel poverty, it became an issue
of high political significance immediately after the Labour Party won the 1997 election.
The responsible minister at DETR announced that the government would introduce new
policies to tackle fuel poverty, one of which was to reduce VAT on domestic fuel to 5%,
the minimum rate permitted by EU rules (Boardman 2010).
EESoP 2 reflected the new emphasis on fuel poverty, at least partly. Using the same
criteria as EESoP 1, for EESoP 2 OFFER expected that about 67% of the total expenditure
for the scheme would be spent in the low-income group (OFFER 1998a). Again, there was
no reference to fuel poverty, but the phrase ‘low-income customers’ was used for the first
time in SO related documents.
In the 1999 consultation paper on energy efficiency OFGEM explicitly mentions fuel
poverty as one of the key objectives of energy efficiency schemes (OFGEM 1999). This
144
Case I: Supplier Obligation in the UK
was also reflected in OF EM’s Social Action Plan (OFGEM 2000c), a document that the
new Labour Government asked the regulators to prepare as a result of a fundamental
review of utility regulation (DTI 1998b). Putting social objectives on utility regulators was
a general trend after the election of Labour in 1997 (Sharratt et al. 2007) and similar
initiatives took place in the water and telecommunications sector (Jones 2001). This
approach was also incorporated legally in the 2000 Utilities Act which states that the
regulator and the Secretary of State must have regard to the interests of (a) individuals who
are disabled or chronically sick, (b) individuals of pensionable age, (c) individuals with
low incomes, and (d) individuals residing in rural areas. As a result, it came as no surprise
that when presenting the framework for EESoP 3, OF EM stressed that the SO ‘can help
to tackle fuel poverty’ (OFGEM 2000b, p. 3).
In addition to the Utilities Act, tackling fuel poverty became enshrined in legislation in
2000 when the Warm Homes and Energy Conservation Act passed through parliament.
The Act required government to develop ‘a strategy setting out the authority’s policies for
ensuring, by means including the taking of measures to ensure the efficient use of energy,
that as far as reasonably practicable persons do not live in fuel poverty’. The Warm Homes
and Energy Conservation Act demanded that a target date must be specified not more than
15 years from the date on which the strategy is published. In 2001, the government
published the UK Fuel Poverty Strategy and committed to eradicate fuel poverty in
vulnerable households in England by 2010 (DTI 2001). This was an interim target with the
long-term target being to eradicate fuel poverty as far as is reasonably practicable by 2016
given the publication date of the strategy in 2001. Because the document did not set an
explicit target for 2016, i.e. the eradication of an unknown number, it was criticised (Hills
2011). The SO is mentioned several times in the document as one of the measures that
would help to bring down fuel poverty.
145
Case I: Supplier Obligation in the UK
Around the same time when the UK Fuel Poverty Strategy had been published, the
government set up a Fuel Poverty Advisory Group (FPAG) for England consisting of
representatives from the energy suppliers, fuel poverty NGOs, consumer and housing
groups, ACE, and independent experts.6 In its first annual report FPAG named the SO as
one of the three key policies to reduce fuel poverty in the UK (FPAG 2003). FPAG also
voiced concerns about the effectiveness of the SO to reduce fuel poverty (for example
because of the likely promotion of light bulbs in fuel-poor households rather than
insulation measures) and demanded a modification of the SO for the purpose of reducing
fuel poverty.
Rising fuel poverty despite government strategy
Ironically, shortly after the adoption of the UK Fuel Poverty Strategy fuel poverty
increased rapidly. The problem with the UK Fuel Poverty Strategy was that it relied on
falling or at least not rising fuel prices (Boardman 2010). However, energy prices
increased significantly over the last decade as pointed out in an earlier section. As a result,
the number of households living in fuel poverty actually went up, not down following the
adoption of the UK Fuel Poverty Strategy.
6
Similar advisory groups were set up in Scotland, Wales, and Northern Ireland.
146
Case I: Supplier Obligation in the UK
Figure 9: Number of households in fuel poverty in England (1996-2009)
Source: based on DECC (2012b) figures for 1997, 1998, and 2000 interpolated
Similar developments can be found in Scotland, Wales, and Northern Ireland (Boardman
2010). This trend was flying in the face of the fact that access to affordable warmth was
one of the four main goals in the 2003 Energy White Paper (DTI 2003) and remained a
goal in the 2007 Energy White Paper (DTI 2007).
In every report produced on fuel poverty in England, FPAG criticised government for not
delivering on fuel poverty and demanded more resources to be put into its eradication
(FPAG 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011). The anticipated failure of
eradicating fuel poverty in vulnerable households led to a judicial review requested by
Friends of the Earth and Help the Aged. However, the judicial review resulted in a ruling
that government had done its job by publishing the Fuel Poverty Strategy and setting a
target date as required by the Warm Homes and Energy Conservation Act. This was
despite the fact ‘that the strategy might be inadequate, provisional and out of date’
(Boardman 2010, p. 14). In the end, the interim target date passed after ‘a devastating
period for the fuel poor’ (FPAG 2010).
147
Case I: Supplier Obligation in the UK
Tension of objectives of SO
Due to the rising number of households in fuel poverty there was constant pressure on the
SO to address the issue at least to some extent. This caused a lot of tension due to the
different objectives of the SO (reducing carbon emissions on the one hand and fuel poverty
on the other), a process that is not unusual when putting social and environmental
obligations on private utilities in a liberalised market (Jones 2001).
There is an incentive for suppliers to work mainly with well-off customers because they
are able to make a higher contribution to the measures delivered (Interview UK4, UK5,
UK6, UK7) and suppliers may get business out of engaging with them whereas lowincome customers are perceived as being less attractive financially (Interview UK3, UK7).
By putting in place a PG target, suppliers are forced to allocate a large share of their
activities to those assumed to be vulnerable.
According to the consultation document on EEC 1, alleviating fuel poverty was one of the
three main objectives of the scheme (after reducing carbon emissions and lowering the
costs of energy). This is also mirrored in the introduction of the obligatory PG target of
50%; previously there was only an indicative target (DEFRA 2001). The introduction of a
firm PG target is clearly related to the Labour Government and its election in 1997
(Interview UK1). The logic of the PG is that all households contribute to the SO with their
energy bills and therefore all households, and particularly the fuel poor, should benefit
from the SO. Ideally, this would make the benefit allocation progressive (Eyre 2008).
But there are some trade-offs:

Regressiveness: As described above, the costs of the SO to suppliers are passed on
to customers in the form of increased energy bills. By definition, the revenue
raising part of the SO is therefore regressive as is any other policy paid for via
148
Case I: Supplier Obligation in the UK
energy bills (Boardman 1993, 1995; Dresner and Ekins 2006; Ekins and Lockwood
2011). Due to confidentiality issues, no public data on the exact costs is available.
Suppliers pass on the cost at their own discretion (Boardman 2004) and are likely
to spread the cost unevenly across customers putting the burden on ‘non-switchers’
and those not on direct debit (Preston et al. 2010), a group that is overrepresented
amongst fuel poor households (Boardman 2010). With increasing size, projected
contributions from customers to the SO went up from just £1 per customer per
annum in 1994-98 (OFFER 1994) to about £54 in 2012 (including both CERT and
CESP) (OFGEM 2012b), although ex-post estimates for EEC 1 and 2 show that
suppliers delivered the obligation for 20% and 23% less than expected,
respectively. Even though the cost effectiveness of the obligations increased
substantially over time (Lees 2006, 2008), rising energy prices are regressive in
that they hit families in the lower income bracket proportionally harder than those
with higher incomes. Hence, if insufficient numbers of households in fuel poverty
are beneficiaries, increasing reduction effort via the SO has regressive impacts.

Cost effectiveness: The higher the fuel poverty component is, the higher are also
the costs of delivery, first because PG households usually receive the measures for
free, i.e. with a 100% subsidy compared to for example, a 50% subsidy in the nonPG, and second because it is more difficult to identify people who are willing to
receive energy efficiency measures when they have to be within a defined group.
This effect shows in evaluations of past SO schemes. An example is the Energy
Efficiency Commitment 2005-2008, which estimates that suppliers spent more than
60% on delivering the PG obligation even though it was only 50% of the overall
target (Lees 2008). As this does not include the indirect cost, the figure is likely to
be even higher. The introduction of the SPG led to higher supplier search costs
149
Case I: Supplier Obligation in the UK
resulting from sourcing the same share of carbon savings in a smaller pool of
household opportunities (Interview UK2, UK7).

Rebound effect: Low-income households take more of the energy savings in the
form of increased comfort (rebound effect) because they tend to underheat their
homes compared to the able-to-pay (Milne and Boardman 2000; Sanders and
Phillipson 2006). Obviously, providing people with the means to adequately heat
their homes has many well-documented benefits including improved health and
well-being (Boardman 2010) and justifies policy intervention before even
considering the impact on carbon abatement. However, due to the increased
comfort taking of low-income households, it is also clear that in order to achieve
the same carbon reduction, a greater energy efficiency improvement associated
with higher costs is required compared to the able-to-pay. One interviewee
representing Consumer Focus, a consumer organisation, claimed that for this
reason early versions of the SO tended to focus on the able-to-pay (Interview
UK14).
These trade-offs were subject to controversial debates over the last ten years and remain
largely unresolved. FPAG was supportive of using the SO as a mechanism to address fuel
poverty, even though they would have preferred a grant scheme such as Warm Front to
take a more important role:
Ideally, any additional resources needed to meet the fuel poverty targets should, in the Group’s
view, be provided through fuel poverty programmes like Warm Front. EEC could then be
focused more on carbon savings. However, assuming a limitation on public resources, EEC
would need to play a significant role in meeting fuel poverty targets as well as helping lowincome customers more broadly on equity grounds. (FPAG 2004, p. 10)
150
Case I: Supplier Obligation in the UK
Subsequently, the SO was seen as a programme that could do both, reducing carbon
emissions and contributing to the eradication of fuel poverty. However, at the latest from
EEC 2 onwards it became clear that there was some confusion over the goals of the SO
and also some evolving conflicts. While eradicating fuel poverty is mentioned in all the
consultation documents following EESoP 3 as one of the objectives of the SO, the primary
aim of the SO remained the reduction of carbon emissions. This becomes very clear in the
EEC
consultation document. The document stresses that EEC 1 ‘was not intended to
specifically target the fuel poor’ (DEFRA 2004a, p. 7) and that the ’primary aim is to make
a significant contribution to the UK’s legally binding target under the Kyoto protocol’
(ibid., p. 5). This is restated more strongly in the CERT consultation where it is stressed
that the SO ‘does not have a specific fuel poverty objective’, that the P
target was put in
place ‘for reasons of equity’ (DEFRA 2006a, p. 7) and that the SO will ‘only ever be able
to be a make a limited contribution to meeting our fuel poverty targets’ (ibid., p. 29). The
change in focus has been described as ‘a discursive shift away from fuel poverty, toward a
more ‘pure’ carbon reduction market’ (Powells 2009, p. 2352) and one interviewee argued
that ‘the move to push for fuel poverty from these energy efficiency programmes [SO] has
been slightly undermined over the time’ (Interview UK5).
In contrast to DEFRA’s consultation documents, FPA
sees the SO as a ‘fuel poverty
programme’ (FPAG 2005), a view opposed by a former director of a UK energy agency:
[…] it is not a fuel poverty target, had never been a fuel poverty target. This is an
environmental programme. And the Priority Group is because it is a social equity issue. […]
Don’t try and steal money from an environmental programme, make your programme work.
(Interview UK4)
Consumer groups recognise the fact that initially the provisions for low-income
households were related to social equity rather than fuel poverty (Interview UK14).
151
Case I: Supplier Obligation in the UK
However, FPA
demanded the P
component of the SO be ‘radically reformed’ in order
to sharpen the SO’s fuel poverty focus (FPAG 2006, p. 3) and also that the PG savings
target should be at least 50% because otherwise it would be highly regressive (FPAG
2007). FPAG subsequently criticised the reduction of the PG target from 50% (EEC 2) to
40% (CERT) as well as the enlargement of the PG to all households on benefits and all
those over 70 (FPAG 2008). Because the total size of the SO increased by 100% from EEC
1 to CERT it meant that the number of households benefiting in the PG increased (but also
the contribution from fuel-poor households to the SO), even though the share of the total
target got lowered. FPAG acknowledged that but would have liked the PG target to have
remained at 50% nevertheless.
Effective lobbying by organisations such as Consumer Focus, National Energy Action
(NEA), Age Concern and Help the Aged (now Age UK) put further pressure on DEFRA to
make provisions for fuel poverty (Interview UK3, UK14). NEA wanted the SO to include
‘targets which focus on the achievement of affordable fuel costs’ (NEA 2000, p. 1),
although NEA argued that the ‘unsatisfactory attempt to shoehorn a social dimension into
an energy saving scheme suggests that in the longer term it may be more effective to
separate these objectives’ (NEA 2001, p. 3). Given that this was unlikely to happen, NEA
demanded a PG target of at least 75% for EEC 1 (NEA 2001), 62.5% for EEC 2 (NEA
2004) and at least 50% for CERT (with 65% of resources spent in PG) (NEA 2006).
However, NEA maintained that the ‘different objectives tend to produce sub-optimal
outcomes for both’ (NEA 2007, p. 3) and that a programme funded by taxation would be a
progressive option of financing both carbon reduction and fuel poverty alleviation
activities. When CERT was extended, DECC introduced the SPG in order to focus more
activity on those affected by fuel poverty. NEA supported this move and lobbied for the
SPG target to be higher than the proposed 10% share (NEA 2010). It ended up being set at
152
Case I: Supplier Obligation in the UK
15% of the total target and one interviewee commented that ‘SP was a ‘concession to the
fuel poverty lobby’ (Interview UK7).
The suppliers, on the other side, generally opposed high PG targets and one interviewee
suggested that they ‘would have been happiest with no P
at all’ (Interview UK1). Both
the enlargement of the PG and the drop from 50% to 40% was a result of DEFRA listening
to the concerns of the energy suppliers (Interview UK6). ERA urged government to reduce
the PG target even further to 25% and, in conjunction with NIA, provided evidence on
why a higher target would not be feasible to achieve 40% (ERA 2007). ERA also pointed
out that it did not believe that the SO was an effective instrument for tackling fuel poverty
and that ‘refining a carbon policy instrument to tackle fuel poverty is not the right
approach’ (ERA 2007, p. 2). Instead, ERA argued that a ‘Private Sector Single Scheme’
should be introduced funded by the utilities but run by organisations with expertise in
social welfare provision (ERA 2006). Similar views were put forward by the energy
efficiency industry. ACE stresses that the SO ‘can only coincidentally help people living in
fuel poverty’ and that other mechanisms need to be developed in order to eradicate fuel
poverty (ACE 2006, p. 7).
With energy prices unlikely to fall significantly over the next years and fuel poverty still
being at very high levels, the conflicts sketched above will probably not go away:
So there was a lot of toing and frowning about that and kind of what is and isn’t about the
programme. Is it primarily carbon, is it fuel poverty, or is it both. And I don’t think that issue is
anywhere near resolved. (Interview UK8)
Summary
Fuel poverty policy did have a significant effect on the SO, particularly after the Labour
Government came into power in 1997. While it was not the main driver for the increase in
153
Case I: Supplier Obligation in the UK
the targets,7 the introduction of the PG is clearly related to debates around fuel poverty.
With a scheme being funded by energy bills and incentives for suppliers to engage with the
well-off customers, there is a strong rationale to make provisions on the grounds of social
equity. However, the extent to which this was deemed acceptable varied and the different
stakeholders took diverging views on the matter. The constant tension over the SO’s
objectives, including the resistance of the energy companies to high PG targets and the
preference of the fuel poverty lobby for a separate scheme funded by general taxation,
clearly falls into the category of the negative policy feedback (Weaver 2010). However,
even though such negative policy feedback processes occurred, the SO continued to make
provisions for low-income households because of the lack of other ambitious fuel poverty
policies and accumulating pressure in the form of increasing fuel poverty mainly resulting
from rising energy prices.
4.2.7 Energy efficiency industry
Another driving force that has been identified is the growth of the energy efficiency
industry as a result of the SO. While the cost of the scheme is born by the energy suppliers
and their customers, the beneficiaries of the policy instrument clearly are the companies
providing the energy efficiency measures incentivised by the SO. With expenditures of
currently more than £1 billion per year the SO creates a significant market for the energy
efficiency industry. As one interviewee put it:
7
Some stakeholders voiced concerns about increasing the targets due to the regressive nature of the
SO’s funding basis. For example, NEA (2004) stressed that without sufficient provisions for the fuel
poor, EEC 2 may aggravate fuel poverty rather than reduce it.
154
Case I: Supplier Obligation in the UK
[Y]ou have in effect an entire industry that built up around it. An industry and a market that
doesn’t really exist in its own right. It is a highly subsidised market, therefore it is probably not
really a genuine market. (Interview UK8)
Another interviewee even mentioned the term ‘subsidy junkies’ when asked about the role
of the energy efficiency industry (Interview UK14).
Interest groups which benefit from public policies are a classic example of positive policy
feedback (Béland 2010; Pierson 1993a). It is therefore not surprising that a policy such as
the SO with substantial monetary benefits for a particular industry displays this effect.
Unfortunately, no reliable data of the past growth of the insulation industry in terms of
turnover, number of companies or number of jobs could be identified and the few
assessments that exist are dated (e.g. Element Energy Ltd 2008; Purple Market Research
2007, 2009). Still, the importance of the SO for the energy efficiency industry can also be
illustrated by looking at the uptake of the various measures that were supported by it. A
good example is insulation measures, which are the most important measures in terms of
their contribution to the overall targets.
Growth of insulation industry
While at a very low level initially, primarily because of the low targets in the beginning
and also because EESoP 1 and EESoP 2 focused solely on homes heated by electricity, the
number of insulation measures supported by the SO increased rapidly particularly after
EEC 1 started. The following graph shows data on the installation rates of cavity wall, loft,
and solid wall insulation from 1994-2010.
Note that the data refers to installations per year based on published OFGEM reports and
they exclude any carry-over activity from previous schemes in order to represent actual
155
Case I: Supplier Obligation in the UK
installation rates. For EEC 1 and EEC 2 no annual figures for solid wall insulation were
available and the total for the whole period of the scheme was allocated on a pro-rata basis
to generate annual figures.
Figure 10: Uptake of insulation measures under the SO (1994-2010)
Source: based on various sources (OFGEM 2005, 2006, 2007, 2008c, 2009c, 2010b, 2011b; OFGEM and
Energy Saving Trust 2003)
Obviously there had been some activity in the insulation market already prior to the SO,
but installation rates were much lower compared to what was undertaken during the last
SO schemes. Before EESoP 1 started in 1994, only around 100,000 cavity wall insulation
installations were carried out per year (Lees 2006). During EEC 2 and the first three years
of CERT more than 550,000 cavity wall insulation installations were carried out on
average every year. Even if one assumes that the activity that had been there prior to the
SO already is included to 100% in that figure, it still means that the market for cavity wall
insulation effectively grew by a factor of about five within just ten years as a result of the
SO. Figures on the business as usual installation rates are also provided by the Impact
Assessment of the Green Deal and the Energy Company Obligation which indicate that
less than 50,000 cavity walls would be insulated without policy support (DECC 2011d).
Those figures imply that EEC 2 and CERT increased the installation rate by a factor of ten.
156
Case I: Supplier Obligation in the UK
Energy efficiency lobby
According to Toke (2000), the interest groups around energy efficiency did not have much
influence in the 1980s and early 1990s, but at the end of the 1990s they were in a powerful
position in the decision making process.
Particularly over the last 10 years the energy efficiency industry has been getting stronger
and continuously put pressure on government to increase the savings targets (Interview
UK3, UK8, UK12). One interviewee mentioned the 68% insulation target introduced as
part of the CERT extension in 2010 as proof of the influence the insulation industry now
exerts:
I think over the years as they become bigger they become more organised, they become more
professional, they’ve got the trade associations the NIA [National Insulation Association] and
ACE which both are quite well connected politically. And you can see that in the CERT
extension you’ve got a 68% insulation target. I mean that has come directly from the insulation
industry saying you need to keep increasing the target, there is more in our supply chain to do
this at least. (Interview UK12)
ACE has been lobbying governments since 1981 (Pickvance 2009), and while the
organisation was formed with an initial membership of only three companies (Dawkins
1987), it now represents twenty companies with a substantial interest in the UK energy
efficiency market. Members include insulation manufacturers such as Knauf Insulation
(UK) Ltd.; glass manufacturers such as Pilkington plc; controls manufacturers, e.g.
Honeywell Control Systems Ltd.; installers of energy efficient equipment such as Mark
Group; but also two major energy suppliers: E.ON UK and ScottishPower. In the early
days of its existence ACE had limited impact on government policy, although it quickly
157
Case I: Supplier Obligation in the UK
developed a visible public and press profile (Owen 1995) and reached ‘insider status’
(Dawkins 1987, p. 255).
Andrew Warren, director of ACE, was a Special Advisor to the House of Commons Select
Committee on the Environment and has given evidence to many House of Commons and
House of Lords Select Committees. ACE also frequently responds to government
consultations (more than 120 from 2001-2011) and asks for a long-term policy framework
and energy efficiency targets (e.g. ACE 2002). In its responses ACE argued for higher
energy savings targets; for example, ACE demanded that CERT should be at least twice
the size of EEC
if not bigger in order to ‘encourage business to maintain current
capacity’ (ACE 2006, p. 7). The association highlighted that 70% of the overall EEC 2
target had been achieved after only 15 months, leaving 21 months for the remaining 30%.
In order to keep the order books of the energy efficiency industry full, ACE argued, a
much higher target for CERT was needed. The same argument was used in a response to a
consultation on the 20% CERT uplift in 2009: ACE stressed that after just nine months
almost half of the CERT target had already been reached and that the uplift could be
increased further (ACE 2009). Concerns raised in the DECC consultation document
(DECC 2009a) about the capability of the insulation industry to deliver at a higher rate
were disregarded by ACE in a response to the consultation on the extension of CERT to
December 2012 (ACE 2010).
The stop-start issues around the SO clearly were a major driver and had an impact on
government thinking:
But as time went on and they got more experience of doing it and each time the targets were
delivered before the end of the period so government then would have to think about the
insulation industry or the lighting industry, you’ve got to have to signal what’s going to happen
158
Case I: Supplier Obligation in the UK
in the next period, you’ve got to make it bigger and to carry over so it doesn’t stop. That kind of
really paved the way how each time it got extended, become widened and become bigger really.
(Interview UK12)
While not as visible as ACE, NIA is another important energy efficiency industry lobby
organisation. NIA (called National Cavity Insulation Association until 2002) has operated
since 1982. It represents more than 170 manufacturers and installers of cavity wall
insulation, loft insulation, draught proofing, external wall insulation and other energy
efficiency products in the UK. Among its members is also one major energy supplier,
E.ON UK. Like ACE, NIA frequently provides evidence to Select Committees and
responds to relevant consultations expressing its support of the SO, asking for higher
targets and highlighting the capability of the insulation industry to deliver more. NIA also
refer to the stop-start pattern of the SO delivery pointing out that this is a real concern for
the insulation industry (e.g. 2006, 2007; NIA 2009).
There are many other lobby groups representing the interests of the energy efficiency
industry including the Energy Services and Technology Association, the Mineral Wool
Insulation Manufacturers Association, the Thermal Insulation Contractors Association, the
External Wall Insulation Association, the Heating and Hot Water Industry Council, the
British Electrotechnical and Allied Manufacturers Association, the Heat Pump
Association, the Combined Heat & Power Association, the British Glass Manufacturers'
Confederation, the Glass and Glazing Federation, and the Draught Proofing Advisory
Association. This list is by no means comprehensive, but gives an indication of the number
of lobby groups acting on behalf of the energy efficiency industry.
An umbrella body, the British Energy Efficiency Federation (BEEF), acts on behalf of 19
trade associations involved with the energy efficiency market. Its chairman is Andrew
159
Case I: Supplier Obligation in the UK
Warren. BEEF was founded by an initiative of the British government in 1996 to ‘provide
a forum for consultation between the relevant trade associations within the energy
efficiency industry and
overnment’ (BEEF 2002, p. 1). The Federation meets quarterly
with government officials to discuss relevant matters. According to Toke (2000), BEEF
helped the energy efficiency industry to gain increased access to civil servants and
ministers by, for example, being involved already at the pre-consultation stage rather than
after proposals are finalised.
Energy Efficiency Partnership for Homes
BEEF also played a major role for setting up the Energy Efficiency Partnership for Homes
(EEPH), which is an important forum for the concerns of the energy efficiency industry. It
was launched by Michael Meacher, the then Minister of State for the Environment, in 2000
and is a network of currently over 600 organisations from the public, private and voluntary
sectors involved in the delivery of energy efficiency measures in the domestic sector
including the big six energy suppliers. According to EEPH, the idea of founding such a
network came out of discussions between DETR, EST, and BEEF after the UK signed the
Kyoto agreement. It has been facilitated by EST.
According to EEPH, the forum ‘enable[s] supply chain stakeholder collaboration which
enhances the evidence base that underpins government’ and its work ‘results in the
identification of opportunities which improve the ability of the industries to deliver on
government goals and targets’ (EEPH 2010, p. 6). Its membership base grew from about
200 organisations in 2000 to more than 600 in 2011 and EEPH can probably be considered
as the most important forum in the area of home energy efficiency policy in the UK
(Interview UK4).
160
Case I: Supplier Obligation in the UK
EEPH is now widely used by government to consult with stakeholders on home energy
efficiency policy. The impact of EEPH on the development of the SO was also highlighted
by a senior civil servant at DECC:
[…] [W]e used the Energy Efficiency Partnership for Homes which is one of their [EST’s]
biggest stakeholder groups to in a sense to sort of stress test the different assumptions
underpinning the illustrative mix. (Interview UK2)
Following Toke’s observation that BEEF provided the energy efficiency industry lobby
with opportunities to influence the policy process pre-consultation stage, EEPH took this
one step further. Interestingly, Andrew Warren, the chairman of BEEF and director of
ACE is also one of the two deputy directors of EEPH. It seems that ACE is extremely well
connected which was confirmed by one interviewee who, when asked which were the
main actors in the policy process prior to EEC 1, answered: ‘ACE obviously’ (Interview
UK4). Although ACE only employs 11 staff it is one of the key players in home energy
efficiency policy in the UK, a fact that had also been noticed by Owen (1995) at a time
when ACE employed just 5 staff.
Summary
The energy efficiency industry undoubtedly benefited significantly from the SO and
markets such as the cavity wall insulation market grew substantially over the last decade as
a result of EEC 1 and 2 as well as CERT. Hand in hand with a growing industry also goes
a strengthening of its lobbying power. While it is not possible to identify the particular
features that were modified in the SO as a result of pressure by organisations such as ACE,
NIA, and BEEF, there is evidence that those groups managed to get involved more directly
in the decision making processes, for example by being consulted prior to the official
consultation stage and being part of important stakeholder forums such as EEPH. The
161
Case I: Supplier Obligation in the UK
expansion of energy efficiency lobby groups clearly falls into the category of positive
policy feedback, contributing not only to the maintenance of the SO, but also to its
expansion.
4.2.8 Perceived success of policy instrument
Growing experience with policy instruments usually results in the expansion of the
administrative capabilities of the state and its agencies making it more likely that similar
policies are implemented in the future (Pierson 1993a; Skocpol 1992). The same is true for
the regulated entities which equally reorganise according to a particular policy regime.
Once established, these effects contribute to a positive perception of a policy, which has
also been a driving factor of the SO. Note this does not refer to the ‘real’ effectiveness of
the policy instrument (e.g. how many tonnes of carbon emissions were saved), but to how
it is perceived by policymakers and other stakeholders. While initially there were concerns
about how such a policy instrument might work and what the costs and benefits would be,
government became much more confident in the SO and considered it as a very successful
instrument (Interview UK1, UK2, UK3, UK4, UK5, UK8, UK12). This also comes out of
the government reports in which the SO was labelled as ‘highly successful in delivering
cost-effective energy efficiency improvements’ (DEFRA 2007b, p. 28), the ‘flagship
energy efficiency scheme’ (DECC 2009b, p. 5), and ‘one of the most cost-effective
policies to reduce carbon emissions’ (DEFRA 2007a, p. 4).
Initial concerns
It did take a while until the SO was so highly regarded. At first, people did not want to
over-commit to something they were not familiar with and there were concerns about the
costs and the benefits, and whether it would work (Interview UK2, UK3, UK9). However,
those concerns diminished with time and ‘people felt more confident, it seemed to be
162
Case I: Supplier Obligation in the UK
pretty successful, it seemed to be popular, and then they were prepared to push on a bit
further’ (Interview UK3). Part of addressing those concerns was the publication of an
assessment by the National Audit Office in 1998 which concluded that EESoP 1 was a
successful and cost-effective scheme that should be extended and widened in the future
(National Audit Office 1998). Also according to the results of the Climate Change
Programme Review, the Energy Efficiency Commitment was one of the most costeffective policies to reduce carbon emissions (DEFRA 2006b). In its CERT consultation
DEFRA announced that it had therefore been ‘decided to maximise carbon emission
reductions via this policy mechanism’ (DEFRA 2007a, p. 4).
Role of suppliers
In an early paper about the politics of energy efficiency and the role of ACE, Dawkins
(1987, p. 263) predicts that in case of the implementation of a policy such as the SO in
reat Britain ‘we would probably see the massed ranks of the energy supply lobbies
coming into the fray’. Resistance from vested interests constitutes a well-recognised
phenomenon (Baumgartner 2009). However, when the SO was first discussed in the early
1990s, resistance was not as fierce as predicted. British Gas openly opposed a policy
similar to the SO but the electric utilities reacted in a more measured way with some
utilities such as East Midland Electricity and Manweb even supporting a new policy
framework carrying out pilot projects in that area. Other electricity utilities, such as South
Western Electricity and Eastern Electricity, were more concerned and reluctant to embrace
the concept of utility energy efficiency programmes (House of Commons Environment
Committee 1993a). After implementation and throughout the various schemes, energy
suppliers frequently raised concerns over the deliverability of the SO (Interview UK5,
UK11) and at some stages even, for example prior to EEC 1, that those could never be met
163
Case I: Supplier Obligation in the UK
(Interview UK4). There is also evidence on this from the consultation responses of the
Energy Retail Association (ERA, now Energy UK), which represents the collective views
of the big six energy suppliers. During the EEC 2 consultation ERA raised concerns about
the lack of capacity in the insulation industry to deliver measures and a lack of demand. It
therefore urged DEFRA to ‘come forward with more reasonable target savings’ (ERA
2004a, p. 7). ERA suggested a target about 25% smaller than the proposed EEC 2 target
and argued that just over half of the assumed number of cavity wall insulation installations
‘might be achievable’ (ERA 2004b, p. 14). Similar concerns were put forward during the
CERT consultation in
7 and ERA pointed out that according to ERA’s own analysis
DEFRA had overestimated the volume of CERT targets that could be achieved (ERA
2007). Similar arguments were made by individual energy suppliers in their response to
various consultations. For example, SSE plc (formerly Scottish and Southern Energy plc)
stressed in its response to the CERT consultation that the proposed CERT target would
‘represent an immense challenge for all energy suppliers’ (SSE 2007) and Scottish Power
were concerned that the CERT target would be ‘a significant challenge to deliver and will
put strain on the insulation industry’ (ScottishPower 2007).
However, in the end suppliers delivered and discharged their obligations long before the
end of all schemes (Interview UK , UK5, UK6, UK11, UK1 ); they ‘jumped as high as
the government wanted them to jump’ (Interview UK5). Because suppliers met their
targets so easily, government became more confident that the targets could be increased
(Interview UK2). This is also highlighted in the CERT consultation document (DEFRA
2006a). While energy suppliers kept raising concerns about the size of the targets they lost
credibility given that in spite of pointing out how difficult those targets were they always
delivered what was expected. One interviewee suggested that the companies cried wolf too
often resulting in government becoming more doubtful about their claims (Interview
164
Case I: Supplier Obligation in the UK
UK5). Clearly, the energy suppliers were aware of this and as a senior manager of one of
the big six energy suppliers put it: ‘We have just been victims of our own success.’
(Interview UK12).
When asked whether by consistently over-delivering the energy suppliers did not shoot
themselves in the foot signalling that they can achieve the targets that they deemed not
achievable in the beginning, a representative of an energy supplier mentioned the
reputational risk involved in not discharging their obligation, the potential penalty of up to
10% of global turnover,8 and potential risks of under-delivery if doing it too late in the
scheme (Interview UK12). Suppliers also always delivered their obligation much earlier
than needed because there is a risk that the supply chain struggles with delivering the
measures required if demand is slowing down and increasing shortly before the scheme
concludes (Interview UK2). Furthermore, consumer demand may not be there when it is
needed if companies leave it to the very end (Interview UK6).
Both over-delivery and consistently meeting the targets were a major factor for extending
the SO. In addition, the opposition to the SO from suppliers remained confined to the scale
of the targets and the particular design, such as PG rules. Outright opposition against the
policy instrument was rare, one reason being that the alternatives, such as a levy on energy
and a fund independent of the suppliers, was a less attractive option from an energy
supplier’s perspective:
‘It kind of suits energy suppliers as far as it goes. It is there, they are not fans of regulation
[…]. But from their perspective they’ve got that business established now and it is easy to keep
that ticking along. And they have got to a point now where they see some competitive advantage
8
Note that it is unlikely that the fine would be 10% of global turnover; this is just the upper limit as
one interviewee suggested (interview UK1).
165
Case I: Supplier Obligation in the UK
for them in it in that it gets them in the door, it gets them a profile with the customer, it might
mean that they then can do something to get them onto their tariff at some point. So there is that
side of things, so for them, that suits them. […] And then also going to a potentially more
onerous form of regulation, something that is more administratively burdensome for them is
also not in their advantage. I think there is an element of better the devil you know.’ (Interview
UK8)
Also, the SO does not put suppliers at a competitive disadvantage because all of
them are subject to the same policy. All of the above are reasons for only modest
resistance to specific elements and support of the SO from suppliers.
Summary
While there were initial concerns about the SO and its implications, the policy instrument
is now considered very successful. The main reason for its perceived success is simply that
it worked in the sense that energy suppliers delivered their obligations. Even though
energy suppliers stressed the difficulties of achieving the increasing targets, government
became more and more confident that they would meet future targets based on past
experience. Therefore they felt self-assured to enlarge the targets every time a new SO
scheme was designed. There are a number of processes at work that can be put in more
theoretical terms: First, there was some negative policy feedback when the interest group
of energy suppliers claimed that the SO was very difficult to achieve. However, this kind
of policy feedback proved fairly weak and positive policy feedback in the shape of the
targets being achieved more than outweighed negative policy feedback. Additionally, the
administrative capability and with it the confidence of government to run the SO increased
over time. Again, this falls into the category of positive policy feedback.
166
Case I: Supplier Obligation in the UK
4.2.9 Policy drivers not found relevant
EU policy was not mentioned as a driver by any of the interviewees, rather UK policy
makers uploaded the SO as a suitable instrument onto the EU energy efficiency agenda
(see draft of Energy Efficiency Directive which explicitly demands all member states to
implement energy savings obligations although other measures are permitted if amounting
to the same level of energy savings).
Conclusions
This chapter sketched the development of the principal home energy efficiency policy
instrument in the UK from its inception identifying some, but not all, driving forces that
impacted on policy change. The SO shows remarkable and frequent changes in its
development, and arguably few people would have expected the scheme to reach the scale
it arrived at in recent years. The analysis of the drivers shows that it was mainly gradual
processes that caused the changes. This is contrary to the perception that significant policy
change is mainly triggered by crisis events.
The SO was initially driven by attempts to incentivise the efficient consumption of energy
at a time of market liberalisation and constrained public funding. Using utilities as a
vehicle to promote energy efficiency had a high ideational fit with the dominant regulatory
framework at the time. Later on, other issues such as climate change, rising energy prices,
and fuel poverty became more and more important. In addition, changes in institutional
structures and key personnel had a significant impact on the SO. Both were a source of
friction and acted as a bottleneck. The growth of the energy efficiency industry and the
perceived successfulness of the SO put further pressure on the government to stick with
the instrument and increase the targets. Both drivers fall into the category of positive
policy feedback.
167
Case I: Supplier Obligation in the UK
The question remains whether the trend of ever increasing targets will continue in the
future given that rising targets also mean a higher contribution by households on their bills
particularly because the potential for low cost measures will at some point be depleted.
According to a recent government consultation, the successor of the current SO will set a
target equivalent to expenditure levels slightly higher than those generated at the moment.
However, the carbon target is supposed to be much lower due to a redirection of the SO
from low-cost measures to more high-cost measures and the introduction of new policy
instruments to incentivise low-cost measures (Rosenow and Eyre 2012). This could have a
significant effect on the SO once again. Therefore, future developments promise to be an
interesting area for more research.
168
Case II: CO2 Buildings Rehabilitation Programme in Germany
5 Case II: CO2 Buildings Rehabilitation Programme in
Germany
Politics is a slow drilling of hard boards with both passion and measure.
Max Weber, 1919 (quoted in Ringer 2004, p. 249)
While the history of the KfW CO2 Buildings Rehabilitation Programme (CBRP) is seven
years shorter than that of the SO, it is no less colourful and interesting. Some issues were
similar, for example the impact of climate change policy on energy efficiency policy.
Others were distinct, such as the fiscal dynamics. Logically, the structure is similar to the
previous chapter: First, in a descriptive section background information on the CBRP and
the main policy changes that occurred is provided. This is followed by an in-depth analysis
of the CBRP.
5.1 Background and main changes of CBRP
The CBRP is run by the Kreditanstalt für Wiederaufbau (KfW), meaning Reconstruction
Credit Institute. KfW was formed in 1948 after World War II as part of the Marshall Plan.
Since its creation the KfW has run several loan and grant programmes related to housing
refurbishment. The first programmes started in 1990, although their primary focus was not
energy efficiency but modernising the housing stock in the east of Germany after
reunification. Only with the CO2-Minimisation Programme did the KfW introduce a
programme with the specific aim to reduce carbon emissions from the housing stock
(Schroeder et al. 2011).
A summary table of those programmes is provided below.
169
Case II: CO2 Buildings Rehabilitation Programme in Germany
Table 6: KfW housing refurbishment programmes
Name of programme
Housing Modernisation Programme East I
Housing Modernisation Programme East II
Housing Modernisation Programme 2003
Housing Modernisation Programme 2005
CO2-Minimisation Programme
CO2-Building Rehabilitation Programme
Period
1990-2000
2000-2002
2003-2004
2005-2011
1996-2004
2001-today
Source: created by author based on BMVBS (2011)
By far the most significant loan and grant programme in terms of energy savings was the
CBRP which is still running today. While this thesis focuses on the CBRP only, a short
overview of the other programmes is provided as part of the analysis.
Note that there were other programmes until 2010 focusing on non-domestic communal
buildings such as schools, child care facilities, and facilities of organisations with a
communal and social focus. These programmes include the Municipality Loans (20072009), Energy Efficient Refurbishment – Municipalities (2009-present), and Social
Investment – Energy Efficient Refurbishment (2009-present) (Clausnitzer et al. 2011).
However, the focus of the following analysis lies on the CBRP only.
5.1.1 Basic architecture of the CBRP
The CBRP provides low-interest loans and grants to households for specified energy
efficiency measures. In the beginning the CBRP focused only on existing homes but later
on loans and grants were also provided as part of the programme for energy efficient new
buildings. The Federal Government funds the scheme and enables the KfW to issue loans
with an interest rate lower than market rates. In addition, some of the funding provided is
used to issue grants. Making use of both federal funding and national as well as
international capital markets, KfW offers financial products to finance housing
refurbishment. Homeowners, housing companies, and public bodies can apply for loans
and grants at an intermediary bank which assesses the individual financial circumstances
170
Case II: CO2 Buildings Rehabilitation Programme in Germany
of the applicant. The intermediary bank then forwards the application to the KfW that then
approves the loan or grant. Figure 11 outlines the model described above.
Figure 11: KfW schemes and financial arrangement
Source: created by author
The CBRP started in 2001 and is still running. It has become the most important
instrument in Germany to tackle carbon emissions from existing homes. While the
Programme changed over time, the core idea of providing low-interest loans (and later
grants) remained the same. The following section sketches the development of the CBRP
from 2001 to 2010 focusing on the key policy changes that took place. This is followed by
an in-depth analysis of the modifications made over time.
5.1.2 Main changes of the CBRP from 2001 to 2010
The history of the CBRP is characterised by a lot of modifications. As before, this section
only describes those changes without analysing the political dynamics behind them – a
171
Case II: CO2 Buildings Rehabilitation Programme in Germany
detailed investigation of the policy processes behind the modifications that occurred
follows in a later section.
The most important changes occurred in the following areas:
Table 7: Main elements of policy change in case of the CBRP
Main elements of policy change
an increasing though varying budget for the CBRP
the type of measures allowed
the introduction of grants alongside loans
programme restructuring
Those changes will be explained in detail below.
Funding and loans/ grants issued
As already indicated above, the annual funding by the federal Government to support the
CBRP changed significantly over time. There is, however, some confusion over the way
the funding is accounted for. The KfW, the evaluations of the CBRP and the supply chain
use the term CBRP for the loans and grants for domestic energy efficiency retrofits
provided by KfW through the programme called CBRP. The responsible government
department, BMVBS, uses the term CBRP as an umbrella term capturing a wide range of
programmes including loans and grants for municipalities, new buildings etc. In the budget
the term CBRP is used in the same way and funding for the CBRP covers a wide range of
different programmes (KfW, personal communication 15/05/2013). The two definitions of
the CBRP are therefore:

BMVBS definition: CBRP encapsulates loans and grants for energy efficiency
retrofits, loans and grants for energy efficient construction and municipality loans.
172
Case II: CO2 Buildings Rehabilitation Programme in Germany

KfW definition: CBRP encompasses only loans and grants for energy efficient
refurbishment of domestic buildings.
This thesis uses the KfW definition. However, neither KfW nor BMVBS provide federal
funding figures for the CBRP as defined by KfW i.e. the loan and grant programme for
domestic energy efficiency retrofits. However, as Figure 12 shows, from 2001-2004 only
loans for retrofit were funded under the umbrella term CBRP as used by BMVBS. We may
assume therefore that the figures for federal funding from 2001 to 2004 for the CBRP refer
to the CBRP as understood by KfW and this thesis. For the years 2005-2007 an
independent evaluation (Kuckshinrichs et al. 2009) provides what the authors call
‘programme costs’ of the CBRP (KfW definition). By using the average ratio of loans and
grants to programme costs the programme costs for the years 2008-2010 can be calculated.
On this basis, Figure 12 shows the federal funding for the CBRP (KfW definition) and the
loans and grants issued from 2001 to 2010.
Figure 12: Federal funding of the CBRP KfW definition & loans and grants issued
Source: based on BMVBS (2011), (Kuckshinrichs et al. 2009) and own calculations
The loans issued by the KfW more or less follow the federal funding and on average the
ratio federal funding (minus funding for grants) to loans issued is around 1:4. The reason
173
Case II: CO2 Buildings Rehabilitation Programme in Germany
for this relatively low ratio is that the interest rate reduction is not only paid in the first
year of the loan but over the whole credit period. Hence the funding paid to KfW covers
the interest rate reduction over the whole credit period for all loans issued in a given year
rather than just the interest rate reduction for all loans that have been issued since the start
of the programme. For example, for a loan of €5 ,
paid back over
years (KfW loans
have credit periods of up to 30 years) with an interest rate reduction of just 1% the total
cost of reducing the interest rate that accrues over time adds up to about €1 ,
this is about €
,
. For %
.
Grants were only issued from 2007 onwards (see below) and remain at a comparably low
level when looking at the scale of loans provided by KfW. However, since the introduction
of grants in 2007 their amount increased more than tenfold by 2010.
Mix of measures
Over time, the mix of eligible measures and the possible combinations of measures
changed. From August 2001 to March 2009 loans provided by the CBRP only funded
packages of measures, single measures were explicitly not supported (Clausnitzer et al.
2010).
At its inception the CBRP only supported the following packages of measures:

Package of measures 1:
- renewal of heating system
- loft insulation
- external solid wall insulation

Package of measures 2:
- renewal of heating system
174
Case II: CO2 Buildings Rehabilitation Programme in Germany
- loft insulation
- basement insulation
- renewal of windows

Package of measures 3:
- renewal of heating system
- heating fuel switching
- renewal of windows
Extra measures from Packages 1-3 could be added to the chosen package, Packages 1-3 are
just minimum requirements.
Further combinations of measures outside of the scope of packages 1-3 were eligible if the
measures were accredited by an approved architect or energy adviser to save at least 40 kg
CO2/m2. The measures eligible for this feature of the scheme are summarised as Package
of measures 4:

Possible measures for Package of measures 4:
- renewal of heating system
- heating fuel switching
- loft insulation
- external solid wall insulation
- basement insulation
- renewal of windows
- ventilation
- ground source heat pumps
- transparent insulation
- photovoltaic
175
Case II: CO2 Buildings Rehabilitation Programme in Germany
- CHP
- micro-renewables
The four Packages eligible for support by the CBRP did not change in the first two years
of the scheme but in January 2003 another package of measures was added to the list
(Diefenbach et al. 2005):

Package of measures 0:
- external solid wall insulation
- loft insulation
- basement insulation
- renewal of windows
Also, two amendments were made to Package 4 in 2003. Financial support was granted if
carbon emissions of more than 35 kg CO2/m2/year or 30 kg CO2/m2/year were saved
(Diefenbach et al. 2005). However, after 2004 measures resulting in a reduction less than
40 kg CO2/m2/year were not supported by the CBRP any longer and were passed on to the
Housing Modernisation Programme in 2005.
In May 2003, two further packages were added with Package of Measures 6 enabling new
buildings to benefit from the CBRP as well (IER and PROGNOS 2004; KfW 2003):

Possible measures for Package 5:
- replacement of stand-alone heating devices using coal, oil, or gas, night
storage heaters, and coal central heating systems with Energy Saving
Ordinance 2002 compliant heating installations
- replacement of oil or gas boilers installed before June 1982 with oil or gas
condensing boilers in combination with solar thermal or other microrenewables
176
Case II: CO2 Buildings Rehabilitation Programme in Germany

Possible measures for Package 6:
- construction or purchase of new buildings compliant with the standard of
the KfW Energiesparhaus 40 (<40 kWh/m2)
- construction or purchase of new buildings compliant with the Passivhaus
standard
However, the two new packages did not feature for long in the CBRP: Package of
Measures 6 was transferred to the new KfW programme ‘Ecological Construction’ from
January 2005 (BAULINKS 2004). At the same time, Package 5 was taken out of the
CBRP and became part of the newly created Housing Modernisation Programme of the
KfW (KfW 2006b).
The following summary table presents the main changes related to the measures supported
by loans from 08/2001 to 03/2009:
177
Case II: CO2 Buildings Rehabilitation Programme in Germany
Table 8: CBRP loan requirements for packages of measures and refurbishment according to new built standard or better from 08/2011-03/2009
08/2001
Age of buildings eligible built before 1978
Package of measures 0 -
Package of measures 1
Package of measures 2
Package of measures 3
Package of measures 4
renewal of heating system
loft insulation
external solid wall insulation
renewal of heating system
loft insulation
basement insulation
renewal of windows
renewal of heating system
heating fuel switching
renewal of windows
requirement to meet <40 kg
01/2003
no changes
external solid wall
insulation
loft insulation
basement insulation
renewal of windows
no changes
05/2003
no changes
no changes
01/2005
no changes
no changes
02/2006
01/2007
built before 1984 built before 1995
no changes
no changes
10/2008
03/2009
no changes
no changes terminated
no changes
no changes
no changes
no changes
no changes terminated
no changes
no changes
no changes
no changes
no changes
no changes terminated
no changes
no changes
no changes
no changes
no changes terminated
no changes
no changes
renewal of heating system
renewal of windows
external solid wall insulation
at least 3 from an energy
expert recommended
measures from:
external solid wall insulation
various other insulation
measures
loft insulation
renewal of windows
ventilation
-
-
terminated
-
-
terminated
as in 2001, but need to as in 2001, but need to reach
CO2/m2 ) with various
measures including insulation
and micro-renewables
reach <35 kg CO2/m2
<30 kg CO2/m2
Package of measures 5
-
-
Package of measures 6
-
-
replacement of stand-alone
no changes
heating devices and boilers
installed before 1982 with Energy
Saving Ordinance 2002 compliant
heating technologies
construction or purchase of new terminated
buildings compliant with the
standard of the KfW
Energiesparhaus 40 (<40
kWh/m2 /year) construction or
purchase of new buildings
compliant with the Passivhaus
standard
Source: based on various sources (Clausnitzer et al. 2010; 2005; Diefenbach et al. 2011; 2006d, 2007)
178
transfer to
Housing
Modernisation
Programme 2005
-
no changes terminated
Case II: CO2 Buildings Rehabilitation Programme in Germany
When the programme structure changed in 2009 (see section below), the requirement to
install a package of measures was lifted entirely and the basic principle became the socalled ‘KfW-Efficiency House’, which defined the energy performance of a refurbished
building compared to a new building according to the Energy Savings Ordinance
(Clausnitzer et al. 2010). The idea to set the standard according to the Energy Savings
Ordinance or a defined benchmark already featured in the CBRP from the beginning
(package of measures 4), but played a less important role. The major change in 2009 was
that this became the main principle of the programme.
Introduction of grants
In addition to the existing low-interest loans, grants were added to the scheme in January
2007. Grants could be used for the same packages of measures and for achieving the
energy performance of a new building or better according to the Energy Savings
Ordinance. Single measures were also supported by grants after 01/2009; the eligible
measures had to be part of the package of measures 4 and the minimum requirements of
the Energy Savings Ordinance needed to be fulfilled. Homeowners could also get a grant
worth 1 % of total investment (maximum €5,
per property) for refurbishment of
existing buildings if they achieved the energy efficiency levels of new buildings. For
refurbishment resulting in a 30% more energy efficient building than new buildings at the
time, 17.5% of the total costs (maximum €8,75 per building) were provided as a grant.
For carrying out packages of measures in existing buildings 5% of costs (maximum € ,5
per property) were paid by a grant (KfW 2006c). However, compared to the loans issued
by KfW, the grants paid were fairly insignificant (in
9: €99 million compared to €3,77
million (BMVBS 2010). As part of the program changes, the threshold for eligible
buildings was raised from 1979 to 1995 (BMU 2007).
179
Case II: CO2 Buildings Rehabilitation Programme in Germany
Programme structure
In the 2007 Integrated Energy and Climate Programme agreed by the Cabinet in Meseberg
(see section below), a substantial restructuring of all KfW programmes related to housing
was announced for spring 2008. In March 2009, the promised restructuring was made
public. The CBRP and the so-called ÖKO-PLUS energy efficiency refurbishment
measures of the Housing Modernisation Program 2005 were merged under the new
umbrella ‘Energy Efficient Refurbishment’. Alongside Energy Efficient Refurbishment the
new programme ‘Energy Efficient Construction’ was created which substituted the former
Ecological Construction programme (BAULINKS 2009). The new structure was supposed
to be more transparent and attractive to households.
The changes also included the start of a funding stream for special measures such as advice
by energy efficiency experts on refurbishment, replacement of storage heaters, and
optimisation of heating systems (KfW 2009).
Other changes
In addition to the changes outlined above there were a number of technical changes in
terms of the minimum requirements of the measures installed. The technical details of the
requirements are, however, outside of the scope of the thesis because they were not subject
to political pressures but rather a result of technology development. An overview of the
changing technical requirements over time can be found in Clausnitzer et al. (2010).
Similarly, the changing financial conditions of the CBRP in terms of the interest rates are
not analysed as part of this thesis. The interest rates are mainly a result of the level of
federal funding and the developments on the financial markets.
180
Case II: CO2 Buildings Rehabilitation Programme in Germany
5.2 Analysis of policy drivers
Following the descriptive section above, this part seeks explanations of the changes that
happened over time. As already done in the UK case study, this chapter uses an approach
that clusters different episodes according to themes rather than reconstructing the history
chronologically.
The following drivers of change have been identified:
Table 9: Main policy drivers of the CBRP
Main drivers of policy change
climate change
struggling construction industry
recession
budgetary considerations
change of key personnel
Of course, these mechanisms are not distinct but interrelated – for example the change of
key personnel (in this case a new Federal Minister who gave a lower priority to building
refurbishment) intersected with debates around the public deficit, arguments that were
used by the minister when amending the budget for the CBRP. Similarly, support for the
struggling construction industry and attempts to alleviate the recession are clearly linked.
However, separate analysis helps to approach this task in a more structured manner than a
chronologic reconstruction would allow.
Similar to the UK case, some of the drivers are more institutional (such as the change of
key personnel) whereas others are more topical (such as climate change policy). Still, no
clear distinction between the two categories can be made because all drivers involve both
181
Case II: CO2 Buildings Rehabilitation Programme in Germany
topical as well as institutional aspects. Note that the drivers covered are by no means
comprehensive and there are other drivers that have not been looked at in detail. At the end
of this section possible additional drivers are discussed. Nevertheless, based on the
empirical evidence reviewed, the above list covers the most important drivers.
Note that the figures for federal funding of the CBRP used in the analysis of policy
changes below refer to the BMVBS definition, as this is the one used in public budgets and
debates. Programme costs for the CBRP KfW definition are somewhat lower and can be
derived from Figure 12.
5.2.1 Climate change
Climate change policy was one of the main drivers for the development of the CBRP
(Interviews G1-G12). At the time of the inception of the CBRP climate change was
already a defined policy area, but this took many years and it is useful to understand where
the debate around climate change was coming from and what role energy efficiency played
in it, particularly with regard to all energy use in buildings.
Beginnings of climate policy in Germany
Until the late 1980s issues around climate change were confined to the scientific
community and political involvement was restricted to funding conferences and research.
Funding for climate research was increased in the 1980s but political interest was limited.
A landmark report was produced eventually in 1985 by the German Physical Society
(DP ) titled ‘Warning of a threatening climate catastrophe’ which received a large amount
of media attention (DPG 1985).
German climate change policy really started with the work of the Enquete Commission
‘Protection of the Earth's Atmosphere’ in 1987. While previously the debate about climate
182
Case II: CO2 Buildings Rehabilitation Programme in Germany
change was largely within the realm of science it appeared for the first time on the political
agenda. The Enquete Commission had the task to review the scientific evidence on climate
change and make policy proposals to Government. In its first report the Commission
concluded that climate change was a serious threat and that political action had to be taken
immediately (Enquete Kommission 1988). This was followed with a recommendation to
reduce CO2 emissions by 30% by 2005 based on 1990 levels (Enquete Kommission 1990).
The Enquete Commission highlighted the importance of energy efficiency for reducing
carbon emissions stating that the ‘Enquete Commission expects the largest realisable
emission cuts over the next two decades to come from the rationale use of energy [my
translation]’ (Enquete Kommission 1990, p. 468) with the biggest potential in the area of
heating in buildings where a 70-90% reduction would be possible. Therefore, the
Commission stressed that policy measures should give priority to those instruments which
lead to an increase in energy efficiency. The recommended measures included a review of
building regulations, funded energy advice programmes, training of architects and
builders, tax breaks for energy efficiency investments, and programmes to provide lowinterest rate loans. This is restated in the final report of the Enquete Commission in which
the Commission proposes a ten year national funding programme for energy efficiency
refurbishments of existing buildings (Enquete Kommission 1994).
Responding to the Enquete Commission’s work, government gave first priority to energy
efficiency, energy conservation, and behaviour change stressing that those could deliver a
20% reduction of CO2-emissions by 2005. In addition to building regulations, government
promised to develop new incentive schemes for energy efficient refurbishments but
remained rather vague about what a scheme like that could look like (Deutscher Bundestag
1993b).
183
Case II: CO2 Buildings Rehabilitation Programme in Germany
First carbon reduction targets
Also in 1990 the German Government promised in its CO2 Reduction Programme to
reduce its CO2 emissions by 25% by 2005 based on 1987 levels, a target that was revised
in the same year with the additional commitment of a significantly higher reduction in new
states. In the following year, this target was revised once again to a 25-30% reduction by
2005 based on 1987 levels but for the whole area of Germany. Because CO2-emissions
were much higher in East Germany at the time, the decision to base the target on the whole
of Germany without sub-targets meant a reduction in ambition and caused political turmoil
(Fleischer 1997).
The work of the Enquete Commission was accompanied by the creation of an
interdepartmental working group on climate change in 1990 composed of representatives
from nine federal ministries plus the Chancellery and the Foreign Office headed by the
Federal Environment Ministry (Krück et al. 1999). This working group still operates and is
tasked with developing and delivering climate change policies. In its second report to the
Bundestag the working group stressed that current funding schemes for energy efficiency
in the building stock would not suffice for generating the investment needed if the CO2targets were to be met. Responding to the report, the coalition government named funding
schemes for retrofitting buildings as one of the key measures to reach its CO2-targets and it
announced that the potential for funding energy efficiency measures in existing buildings
would be assessed in conjunction with other policy instruments (Deutscher Bundestag
1993b). However, two years later the government still promised an assessment of
additional programmes and lauded the initiatives by the Länder asking the regional
governments to consider increasing those schemes (Deutscher Bundestag 1994). While
there had been revisions of building regulations and the various energy performance
184
Case II: CO2 Buildings Rehabilitation Programme in Germany
related ordinances, no tangible policy instrument to fund energy efficiency building
refurbishment in West Germany was put forward. The rational was that East Germany
offered a lot of low hanging fruits and therefore programmes should target this area first
(Deutscher Bundestag 1993a).
First KfW programme focuses on East Germany
As a result of unification, various policies were introduced in order to rebuild the
infrastructure in East Germany. Part of this was the first KfW programme for
refurbishment of the housing stock, the Housing Modernisation Programme East 1. While
the Housing Modernisation Programme East is not the focus of this thesis, a brief review
of its history is provided as background given that the CBRP started in the context of
already existing housing refurbishment schemes.
At the time of the inception of the Housing Modernisation Programme East 1, the
condition of the housing stock in East Germany was very poor and many properties
required repair. Housing Modernisation Programme East 1 ran from 1990-2000 and
resulted in the refurbishment of 3.6 million dwellings (Bundesregierung 2003). Lowinterest loans worth more than €
billion were spent as part of the Housing Modernisation
Programme East 1, most of this was used for repairs and fuel switching from coal to gas
and heating oil (BMU 2006). The Housing Modernisation Programme East 1 was followed
by the Housing Modernisation Programme East 2 which ran from 2000-2002 and
concentrated on the more expensive refurbishment measures. Similarly to Housing
Modernisation Programme East 1, only dwellings situated in East Germany were eligible
for funding. During the two years of the programme low-interest loans were made
available by KfW resulting in more than 159,000 dwellings being refurbished
185
Case II: CO2 Buildings Rehabilitation Programme in Germany
(Bundesregierung 2003). Note that energy efficiency measures were just a small part of the
scheme and only 15% of all loans issued were used to finance energy efficiency measures
(IER and PROGNOS 2004). HMP 2 was succeeded by the two year Housing
Modernisation Programme 2003. In contrast to earlier programmes, the Housing
Modernisation Programme 2003 covered all of Germany. Low-interest loans worth €5.5
billion were provided as part of the scheme. A number of measures were supported by the
Housing Modernisation Programme 2003 and only some of them related to energy
efficiency improvements (Eichhammer et al. 2006). From 2005-2010 a fourth programme
ran, called Housing Modernisation, providing almost €15 billion worth of loans (BMVBS
2011).
In summary, the first funding programmes were part of the efforts to upgrade the
infrastructure in the East of Germany post-reunification, climate change only played a
minor role and the positive impact on reducing carbon emissions was more a side effect
rather than the primary aim of the programme (Schroeder et al. 2011).
The predecessor of the CBRP kicks off
The first KfW programme aimed specifically at reducing carbon emissions of the housing
stock, the CO2-Reduction Programme (CRP), only started in 1996, many years after the
government’s promise to establish an additional funding scheme of some kind. The
scheme was running during 1996-2004 and was a measures-based programme. Eligible
measures included insulation (loft, solid wall, window glazing), increased boiler
efficiency, solar thermal, photovoltaic, heat pumps, heat recovery measures, district
heating, and CHP. KfW provided low-interest loans to home owners, housing companies,
councils, the Kreise (regional government bodies), and other public bodies. At first the
186
Case II: CO2 Buildings Rehabilitation Programme in Germany
scheme applied only to existing buildings but it was opened for the construction of new
low-energy buildings in 1998 (Kleemann et al. 2003). While in the beginning CRP covered
only the old Länder in West Germany, in 1998 the scope was widened and new build lowenergy homes in East Germany were eligible for support. From March 2000, all funding
streams of the CRP were available to the whole of Germany (Deutscher Bundestag 2007).
Loans worth 1 billion DM (about € .5 billion) in the first years of the scheme were funded
by the Federal Government but in later years the loans were provided by KfW without any
further federal funding (Deutscher Bundestag 1997a, 2004).
Climate change climbing up the agenda
After 2000 climate change issues received increasing media coverage. Using the
Süddeutsche Zeitung as a proxy for media coverage of climate change, Schäfer et al.
(2011) show that the proportion of articles dealing with climate change rose significantly
after 2000 and particularly after 2006:
Figure 13: Newspaper coverage of climate change in Germany
Source: Schäfer et al. (2011)
187
Case II: CO2 Buildings Rehabilitation Programme in Germany
Increasing media coverage was accompanied by political commitments to reduce GHG
emissions: in 1995, the carbon emissions target was changed to a 25% reduction by 2005,
but based on 1990 levels as common practice in the international arena (Fleischer 1997).
Under the 1997 Kyoto Protocol Germany committed itself to a 21% reduction in GHG
emissions in the period 2008 to 2012 compared with 1990 levels (BMU 2010), a target
which was less ambitious than
ermany’s domestic target. The domestic target was
renewed in the 2000 National Climate Change Programme and remained at a 25%
reduction by 2005 based on 1990 levels (BMU 2000). The 2000 National Climate
Protection Programme was also the start of the CBRP.
Inception of the CBRP
Being part of the newly established National Climate Protection Programme the CBRP’s
main purpose was also mirrored in its name: CO2 Building Rehabilitation Programme.
The Federal Government provided 400 million DM for three consecutive years, i.e. a total
of 1.2 billion DM, enabling the CBRP to run until 2003. The funding for the CBRP was
based on the revenues from auctioning the Universal Mobile Telecommunications System
(UMTS) licenses which generated more than €5 billion. All of the revenues were used to
reduce the national debt which in turn led to lower interest rate payments (about € .6
billion per year). Part of this reduction of interest rate payments provided the financial
resources for the CBRP in the beginning of the programme (Deutscher Bundestag 2002).
It was also announced that a decision about an extension of the scheme lasting at least two
years would be made in 2003 when putting together the 2004 budget. The projected CO2reduction was expected to be around 5-7 million t by 2005 compared to 1990 levels
188
Case II: CO2 Buildings Rehabilitation Programme in Germany
making the CBRP the most important feature of the package of measures addressing
emissions of households (Bundesregierung 2000).
First extension of the CBRP
In May 2003, Government announced the promised extension of the CBRP to 2005. The
extension also provided an additional €16 million annually taken out of the tax revenue of
the ecological tax.9 This was a significant increase in funding and equated to about 80%
more annual funding for the programme. The additional funds and the extension of the
CBRP were presented as carbon reduction measures in the context of national climate
policy (BMU 2003). Other aspects such as job creation were clearly secondary in the press
announcement released by BMU, BMVBS, and KfW but were referred to as additional
effects of the CBRP (see more in section on construction industry).
Second extension of the CBRP
In his policy statement in March 2005 Chancellor Schröder announced a further extension
keeping funding levels constant. This extension of the CBRP also featured in the 2005
National Climate Protection Programme where it was presented once more as a
programme to save carbon emissions. Because the energy-intensive sector was covered by
the EU Emissions Trading System, the 2005 National Climate Protection Programme
focused on the domestic sector and the transport sector (BMU 2005).
9
The ecological tax reform was introduced in 1999, modified in 2000, and modified again in 2003. The
reform included an incremental increase of taxes on petrol and a tax on gas, oil, and electricity. The
revenue is used to lower the rates paid by employees to the public pension scheme so that the reform
is for the most part revenue-neutral.
189
Case II: CO2 Buildings Rehabilitation Programme in Germany
The 2005 National Climate Protection Programme only stated the Kyoto target and did not
refer to the domestic target as defined in the 2000 National Climate Change Programme
(BMU 2005). However, for the medium- and long-term, the Government promised to cut
GHG emissions by 40% by 2020 based on 1990 levels under the condition that the EU
commits to a 30% reduction over the same period. This ambition was later restated in the
Integrated Energy and Climate Programme (BMU 2007) and in the 2009 coalition contract
of the Christian Democratic Union/Christian Social Union (CDU/CSU) and Free
Democratic Party (FDP) (CDU et al. 2009).
New coalition government increases funding
In September 2005, a new coalition government was elected. The coalition consisting of
the Social Democratic Party (SPD) and the Green Party was succeeded by a coalition led
by the CDU with the SPD as the junior partner. The change in government did not,
however, lead to a significant shift with regard to the CBRP and its priority (Interview
G1). Early in 2006, the new government modified the CBRP once again and an uplift of
the financial resources provided was announced after a Cabinet meeting in Genshagen on a
programme for economic growth and employment. For the period 2006-2009, Government
promised to allocate € billion, a significant increase of the programme’s ambition. The
additional funds more than quadrupled the existing budget for the CBRP. Climate change
did not, however, play a major role at that point – issues around the recession and job
losses in the construction industry dominated the debates. The changes made were clearly
framed as measures to support growth and particularly small and medium-sized companies
(also see sections below) whereas climate change is only mentioned sporadically in the
press release (Bundesregierung 2006) and was not the top priority at the time (Interview
G4, G6). One interviewee, a representative of the CDU, stresses that the increase in
190
Case II: CO2 Buildings Rehabilitation Programme in Germany
funding also had to do with the discussions around the renewable heat law, a policy that
requires new buildings to use a defined proportion of heat from renewable sources that was
part of the coalition agreement (CDU et al. 2005). According to the interviewee, the
controversy around the most sensible order of measures, i.e. efficiency measures in order
to reduce heat consumption prior to renewable heat, led to an increase in the CBRP
(Interview G11).
Meseberg proposals on CBRP
An important framework of German energy and climate policy was the result of a cabinet
meeting in Meseberg in 2007. The cabinet put together the so called Integrated Energy and
Climate Programme, which sets a national target to reduce carbon emissions by 40% by
2020 based on 1990 levels (BMWi and BMU 2007). The Programme contains a list of 29
policy proposals and amendments some of which relate to energy efficiency in buildings.
However, there were controversial discussions around what policy measures should feature
in the Programme. In the run up to the meeting at Meseberg, the Environment Minister,
Sigmar
abriel of the SPD, proposed to increase funding for the CBRP from about €1.
billion to € .5 billion annually. Those plans were opposed by the Federal Ministry of
Finance. Furthermore, Sigmar Gabriel proposed a number of regulations that would have
imposed a number of energy efficiency requirements on home owners such as a ban of
night storage heating systems (Frankfurter Allgemeine Zeitung 2007). One of the
proposals was a requirement to install energy efficient windows in those properties with
non-insulated windows. Also, Gabriel proposed that tenants of energy inefficient
properties should be allowed to pay a lower rent until the landlord upgrades the property to
an acceptable level of energy efficiency (Deutschlandradio Kultur 2007). However, none
191
Case II: CO2 Buildings Rehabilitation Programme in Germany
of those proposals featured in the Integrated Energy and Climate Programme agreed by the
Cabinet in Meseberg. No commitments were made to additional funding for the CBRP, it
was only agreed that the programme would be stabilised at current levels until 2011 and
restructured to make it more effective (BMU 2007; BMWi and BMU 2007). However,
abriel’s proposal to ban night storage systems resulted in grants being offered as part of
the CBRP for the replacement of inefficient heating systems (Interview G3). One policy
measure which was part of the Integrated Energy and Climate Programme was to raise the
requirements for energy efficiency of new buildings in the Energy Savings Ordinance.
This had an indirect effect on the CBRP in the sense that the CBRP increasingly defined
the expected energy efficiency performance in relation to the Energy Savings Ordinance.
The discussions around the CBRP at Meseberg were primarily driven by climate change
considerations (Interview G8), given the focus of the Integrated Energy and Climate
Programme on climate change. In the years following Meseberg, due to budgetary
considerations, the government modified the overall budget of the CBRP. This was not
related to climate change policy; it had more to do with debates around the public deficit
and a change of the Federal Minister of the BMVBS. Both issues are covered in detail
below.
The 2010 Energy Concept
After a new coalition consisting of the CDU and FDP was elected in 2009, the government
published the 2010 Energy Concept, a framework for the future of German energy policy
(comparable to the UK Energy White Papers). The overall objective stated in the Energy
Concept is to transform the German economy to one of the most energy efficient and
environmentally friendly economies in the world, provided that energy prices remain
192
Case II: CO2 Buildings Rehabilitation Programme in Germany
competitive and the level of prosperity stays at a high level. While reducing carbon
emissions was not the only objective of the Energy Concept, it played a major role and
there are many references to climate policy throughout the document. Refurbishing the
building stock is pointed out as one of the key areas with a high potential for reducing
energy demand and carbon emissions and the government commits itself to reaching an
almost carbon neutral building stock by 2050, promising a refurbishment roadmap to begin
in 2020 (BMWi and BMU 2010). However, no modifications of the CBRP were
announced in the Energy Concept and it remained rather vague about the policies that
would be implemented in order to reach the stated policy objectives in the area of home
energy efficiency.
Although not the focus of this thesis, a remarkable policy proposal featured in the original
proposals with regard to obligatory energy efficiency standards for existing buildings
(Interview G2). However, those proposals were watered down after a controversial public
debate and did not feature in the final version of the Energy Concept (BMWi and BMU
2010). Government said in the Energy Concept that it wanted to offer incentives and not
impose forced energy efficient refurbishment on homeowners. Economic considerations
were at the heart of Government thinking, and energy efficiency measures should only be
carried out if they were cost effective. Instead of obligatory refurbishment, the Energy
Concept promised that demolishing and replacing old inefficient buildings would be
covered by the CBRP in the future. In sum, the draft proposals were weakened
considerably, with no precise commitments with regard to the CBRP.
193
Case II: CO2 Buildings Rehabilitation Programme in Germany
Additional funding from the Energy and Climate Fund
Shortly after the Energy Concept had been published, government announced an important
change with regard to the CBRP. After the CBRP’s budget had been cut significantly (see
section on budgetary considerations below), an additional €5
million per year were
allocated to the CBRP. This money was supposed to come out of the Energy and Climate
Fund, special assets outside of the budget (Deutscher Bundestag 2010a). The fund was
created to help finance measures to reach the objectives laid out in the Energy Concept and
has a strong climate component as also reflected in the name. Initially, a tax on windfall
profits of the energy companies due to the prolonged phase-out of nuclear energy was
supposed to provide a substantial proportion of funding for the fund, but after the nuclear
accident at Fukushima, Japan, the German government reversed the plans to prolong the
nuclear phase-out. However, this is beyond the scope of this thesis as those events
occurred after 2010. In addition to the new tax on windfall profits, the revenues of
auctioning the EU ETS permits should generate financial resources for the fund. This was
based on the assumption of a high price per tonne CO2, but prices dropped significantly in
2011 (Bloomberg 2012). Again, this matter is beyond the scope of this thesis and will not
be explored further. The important change that took place is that for the first time part of
the funding for the CBRP was not coming out of the budget but linked to an independent
fund, the implications are discussed in the section on budgetary considerations.
Summary
Climate change policy clearly was the most important driver of the CBRP overall,
although at certain times other issues such as using the Programme as an economic
stimulus played a more important role. However, as one interviewee put it:
194
Case II: CO2 Buildings Rehabilitation Programme in Germany
‘I believe today we have a clear focus on energy efficiency and climate protection. […] The
issues of creating jobs and economic growth are surely a side effect of the programme, I would
not see this as the central motive.’ (Interview G4)
As shown above, this is also reflected in the official government documents and statements
where climate change frequently features as the top priority of the CBRP.
5.2.2 Supporting the construction industry
Although the primary objective of the CBRP was to reduce carbon emissions, the CBRP
always had strong links to policies supporting the struggling construction industry.
References to job creation and stimulating higher turnover in the construction industry can
be found throughout the CBRP’s development and it certainly was a major driver. At
certain times considerations about the construction industry overshadowed climate policy
objectives and there are several examples where the CBRP was changed primarily because
of its effect on employment and growth in the construction sector. This section provides an
analysis of how this policy driver impacted on the CBRP.
Declining turnover and employment in the construction industry
Between 1995 and 2010 the number of jobs in the German construction industry almost
halved with turnover declining by more than 30% in the same period (Statistisches
Bundesamt 2012c). While the trend appears to have changed after 2006, the industry is
still stagnating (Figure 14).
195
Case II: CO2 Buildings Rehabilitation Programme in Germany
Figure 14: Turnover and employment in the German construction industry 1995-2010
Source: based on Statistisches Bundesamt (2012c)
Reasons for the decline of the construction industry are manifold: demographic effects i.e.
a
stagnating population, tailing off of the infrastructure upgrade in East Germany,
competition from abroad, lower demand for new buildings, and decreasing public
investment in infrastructure (Ottnad and Hefele 2006).
In light of this trend, the CBRP was justified from the beginning not just with its
contribution to climate policy objectives, but with the positive effects on the construction
sector (Deutscher Bundestag 2000). These arguments were also used when putting in place
the earlier KfW programmes focusing on building refurbishment (Deutscher Bundestag
1997a, 1997b). However, the primary objective at the time was clearly climate protection
(see section on climate change).
Increasing emphasis of positive impacts on construction industry
As described above, Government extended the CBRP in 2003 to 2005 providing an
additional €16 million annually. Alongside the objective to reduce carbon emissions, the
contribution of the CBRP to job creation and preservation in the construction industry was
196
Case II: CO2 Buildings Rehabilitation Programme in Germany
stressed by both the Federal Construction Minister, Manfred Stolpe, who stated that the
increased budget for the CBRP would secure 18,000 jobs in the construction industry
(Reimer 2003) and Environment Minister, Jürgen Trittin, who claimed that the program
had become a ‘job machine’ (BMU 2003). Not surprisingly, the extension and the increase
in funding was applauded by the construction industry (Bundesvereinigung Bauwirtschaft
2003).
Also when extending the CBRP in March 2005, Chancellor Gerhard Schröder referred to
the need to support jobs in the construction industry particularly in small and medium
enterprises (Bundesregierung 2005b). This was confirmed in the Twenty Point Programme
for Economic
rowth, in which
overnment promised to provide €7
million for the
scheme’s extension, again with a reference to the construction industry alongside with
climate policy (Bundesregierung 2005a).
The government increasingly used arguments with regard to the construction industry and
this became the main reason for extending the programme in 2006 when the new
government modified the CBRP once again. An uplift of the financial resources provided
was announced after a Cabinet meeting in Genshagen as part of a programme for
economic growth and employment. For 2006-
9,
overnment promised to allocate €
billion, a significant increase of the programme’s ambition (Bundesregierung 2006). This
was part of the initiative, ‘Housing, Environment,
rowth’, which featured in the
investment programme announced after the Cabinet meeting. The first of the three reasons
for the uplift named by Construction Minister Wolfgang Tiefensee was job creation and
keeping jobs in the construction industry. In a press release KfW and government argued
that the increase in funding would have immediate positive effects on employment (KfW
2006a). Similar references can be found in later policy documents. Also the emphasis in
197
Case II: CO2 Buildings Rehabilitation Programme in Germany
the paper summarizing the results of the Genshagen cabinet meeting, of which the
extension and top-up of the scheme was part, lies clearly on economic growth and
stimulating additional investments (Bundesregierung 2006). Interviewees also supported
the view that around 2006 job creation in the construction industry played a major role
(Interview G4). Again, the Genshagen proposals were met with support by the
construction industry (Bundesvereinigung Bauwirtschaft 2006).
Asked why the CBRP was picked as the instrument to address the decline of the
construction industry and unemployment in general one interviewee suggested that the
CBRP supports economic activities with a high labour intensity (more than 50%) whereas
other areas have a much lower labour intensity (Interview G1). However, the rhetoric
around job creation and supporting the construction industry appears to be somewhat
instrumental (also interview G12):
‘The CBRP was not created primarily to secure and create jobs, but to contribute to climate
protection, energy security and reduce the cost of energy consumption via demand reduction.
But employment effects are a welcome side effect.’ (Interview G1)
Most interviewees agreed that while the CBRP did have positive effects on the struggling
construction industry and employment the overall aim was related to climate policy goals
(e.g. interview G1, G2, G4, G5, G12).
Summary
Positive effects on the construction industry were always stressed in discussions around
the CBRP. Arguably, other drivers such as climate change dominated in most years. There
were, however, times when arguments with regard to job creation in the construction
industry played a major role. This was most notable during 2005/2006 when both the old
198
Case II: CO2 Buildings Rehabilitation Programme in Germany
SPD-Green government and the CDU/CSU-SPD government justified the increased
funding for the CBRP with its important role for protecting and creating jobs in
construction.
5.2.3 Recession
Another driver that had an impact on the CBRP more recently was the recession triggered
by the global financial crisis in 2008. Aware of the positive effects on economic activity
and employment, the German government used the CBRP as an instrument to stimulate the
economy.
Contracting economy and unprecedented low business confidence
Following the financial crisis, the economic situation in Germany worsened considerably
and the economy contracted by 1.9% in the last quarter of 2008 compared to the previous
year, 6.5% in the first quarter, 7.4% in the second quarter and 5% in the third quarter of
2009 (Statistisches Bundesamt 2012a). Never before since World War II had the German
economy experienced a recession of such scale (Butterwegge 2012).
Business confidence approached the lowest level in decades and the outlook was very
bleak. An influential German index measuring economic activity and the expectations of
businesses is the Ifo Institute’s Business Climate Index. The Index is based on surveys in
manufacturing, construction, wholesale and retail. The survey asks the companies about
their current business situation and their expectations for the next six months. Based on the
survey responses Ifo Institute compiles the Business Climate Index which is published
monthly. The index shows a remarkable dip around 2008 indicating the decline in
economic activity that followed (Figure 15).
199
Case II: CO2 Buildings Rehabilitation Programme in Germany
Figure 15: Ifo Business Climate Index in Germany
Source: based on Ifo Institute (2012)
It was in this context that the government considered potential measures to help stimulate
the economy when the CBRP was modified.
CBRP as a means to stimulate economic activity
There was a debate, at times controversial, about what kind of initiatives the government
should undertake to stimulate economic growth.
The SPD drafted a programme aimed at generating investment worth €6 billion, part of
which was the €3 billion increase in funding of the CBRP that was later incorporated in the
Economic Stimulus Package I (Süddeutsche Zeitung 2008), even though the Finance
Minister at the time, Peer Steinbrück (SPD), positioned himself against such an investment
programme (Frankfurter Allgemeine Zeitung 2008a). In contrast, the CDU/CSU proposed
tax breaks for high income earners and new cars (Focus 2008). However, Chancellor
Angela Merkel was in favour of a non-traditional economic recovery package targeted at
specific sectors of the economy triggering further private investment (Frankfurter
Allgemeine Zeitung 2008b). The CBRP fit into this category and is widely regarded as a
200
Case II: CO2 Buildings Rehabilitation Programme in Germany
programme that generates significant private investment (Kuckshinrichs et al. 2010). For
example in 2009, the ratio of public investments to private investments as a result of the
CBRP was estimated to be 1:9 (BMVBS 2010). Therefore, according to the then
parliamentary group leader of the SPD, Peter Struck, the increase in funding for the CBRP
was one of the measures which was not contested when the coalition government
discussed potential measures (Zylka 2008).
As a result of those discussions, in November 2008 the Government published a first
investment programme as a response to the recession. It was argued that following the
financial crisis the government saw its task in securing growth and employment. Part of
the fifteen point programme ‘Securing Jobs by Strengthening
rowth’ (also known as the
Economic Stimulus Package I) was a promise to top up the CBRP and other KfW
programmes for 2009-
11 by €3 billion (BMWI and BMF 2008). Shortly after the
programme was published, KfW announced an increase of grants for packages of measures
from the previous 5% to 7.5% of total costs (KfW 2008).
While the government always used climate policy related arguments when justifying
previous extensions and funding increases, the document presenting the Economic
Stimulus Package I did not contain a single reference to climate change or environmental
issues. This indicates that at the time the driver for the modifications made was merely the
recession and the perceived need to stimulate economic growth. Such a view is also
confirmed by interviews (Interview G4).
There was a second stimulus package in 2009, but no additional provisions were made for
the CBRP.
201
Case II: CO2 Buildings Rehabilitation Programme in Germany
Summary
Following the financial crisis the CBRP was used as part of the government’s attempts to
stimulate economic growth. While the CBRP’s contribution to economic prosperity was
stressed also prior to 2008, it was only then that the Programme was amended explicitly
aimed at triggering investment to help overcoming the recession.
5.2.4 Budgetary considerations
Another important driver of the CBRP was the debate around public debt and the
implications for public spending. First, this section provides background information on
public debt in Germany and the new constitutional amendment to limit additional public
debt (debt brake). Second, the implications of fiscal policy and budgetary considerations
for the CBRP are analysed in detail. All interviewees agreed that budgetary issues played
an important role for the development of the CBRP (Interview G1-G12).
Growing public debt
While public debt had been rising continuously over the last decades, and the reasons are
manifold following the financial crisis and during the recession it increased at a much
faster pace than in the years before (Figure 16). The last time public debt increased so
quickly in Germany was after reunification in the early 1990s (Streeck 2010).
202
Case II: CO2 Buildings Rehabilitation Programme in Germany
Figure 16: Public debt in Germany as a proportion of GDP
Source: based on Statistisches Bundesamt (2008, 2011)
The main reasons for the rapid increase in public debt as a proportion of GDP around 2008
/2009 are the two economic stimulus packages which required a large amount of public
funding (see section 5.2.3) and the bailout of the banks during the financial crisis
(Statistisches Bundesamt 2010) going hand in hand with lower, and at times even negative,
economic growth leading to a lower tax revenue. Hence the gap between government
revenues and government spending suddenly expanded reaching its peak in 2010 when it
was about seven times bigger than in 2008 (Grasl and König 2010).
Debt brake
Article 115 of the German constitution sets a limit to how much additional net debt the
government can take on every year. From 2002-2006 and again since 2010 the budget did
not comply with Article 115. In order to address the growing public debt, already the redgreen coalition attempted to change the financial constitution as part of the reform of
federalism. However, the Federalism Commission failed to find an agreement due to
controversies around education policy and the different roles of the Länder and the federal
203
Case II: CO2 Buildings Rehabilitation Programme in Germany
government. Following the 2005 election, the CDU and SPD restarted the reform process
and set up a second Federalism Commission in 2006. More than two years later the
Commission concluded and put forward a number of proposals including fiscal policy
reform. For a more detailed discussion of the political processes around the federalism
reform see Lorenz (2010).
Based on the Commission’s proposals the government coalition successfully passed an
amendment of the constitution through the Bundestag and Bundesrat (Federal Council
representing the Länder) (Grasl and König 2010). This amendment limited the amount of
additional debt that the government could take on in the future to just 0.35% of GDP per
year and was called ‘debt brake’. Essentially, the debt brake requires both the Federal
Government and the Länder to balance the budget without additional debt. This rule will
be introduced gradually from 2011 to 2016 and only in emergency situation can the rule be
violated (Renzsch 2010).
There is a connection between the debt brake and the two stimulus economic packages
discussed before: the large amount of public spending and the required additional debt
could only be justified with a commitment to limit the budget deficit in the future (Dietrich
2009).
Budgetary cycle and controversies about variability of funding
Before the debt brake had an impact on the CBRP in 2010, there were discussions about
the amount of funding available. Demand for loans and grants in 2009 was higher than
expected and the funding allocated to the CBRP in the budget was not sufficient to provide
for all applicants. In August
9 federal funding for
9 was raised by an extra €75
million taken from the budgets for 2010 and 2011. Hence total federal funding for 2009
204
Case II: CO2 Buildings Rehabilitation Programme in Germany
was more than € billion (BMVBS 2011). Bringing forward funding from later budgets is
not an unusual process, but this step led to a controversial debate later on.
As a result of the increased spending in
9, only €1.1 billion were left in the budget of
the scheme for 2010. The construction industry association Bundesvereinigung
Bauwirtschaft asked the Government to top up funding for 2011 in order to make sure that
the €1.5 billion per annum announced in the investment package in
8 would be
provided (Bundesvereinigung Bauwirtschaft 2010). The reduced funding in 2010 which
resulted from the shifting of funding from 2010 to 2009 was heavily criticised by various
associations and NGOs. In a campaign the reduced funding was framed as cuts to the
funding of the CBRP (CAMPACT 2010), a claim that is technically not correct given that
the budget was simply spent earlier meaning that for the whole commitment period 20092011 the overall funding did not decrease (Interview G2, G3, G4, G6, G11).
Adding to the controversies, the financial contributions to the KfW programmes were put
on hold in January 2010 and KfW stopped approving applications for loans and grants.
The reason was that the budget for the year 2010 was not yet in place and that demand for
loans and grants was higher in 2009 than expected (EnBauSa 2010c). However, shortly
after the CBRP was put on hold and following objections voiced by various industry
associations and NGOs, the Bundestag approved preliminary spending with regard to the
KfW programmes (EnBauSa 2010b).
There have been calls for more reliable funding over time. However, the budget for such
programmes can only ever be guaranteed for one year as part of the overall federal budget
and the respective laws. A longer commitment over a time span of 5-10 years for example
is not possible within the federal budget and permitted by the Bundesrechnungshof, a
supreme federal authority that examines federal financial management (Interview G6).
205
Case II: CO2 Buildings Rehabilitation Programme in Germany
Debt brake impacts on CBRP
However, in June 2010, a leaked letter to the coalition MPs from the Minister of the
BMVBS, Peter Ramsauer, unveiled that cuts of around 50% of the CBRP were planned for
2011 in addition to the already reduced budget (due to bringing forward funds to 2009)
leaving the scheme with only € 36 million for
11 (Ramsauer 2010). This was later
confirmed in a meeting of the MP committee on transport, construction, and urban
development (EnBauSa 2010d). Reasons given for the cuts were mainly that the debt brake
would not allow for the programme to continue at current levels. Also, it was argued by
the Coalition Government, low-interest rates would enable investors to find loans
elsewhere. Both the SPD (Deutscher Bundestag 2010c) and the Green Party (Deutscher
Bundestag 2010d) put forward a motion to not authorise the cuts. The Green Party asked
for federal funding of the CBRP of at least € billion per year and the SPD demanded
stabilisation at a high funding level. Opposition to the announced changes also came from
all the major associations with an interest in housing (construction industry, housing
corporations, homeowners, tenants) (BAULINKS 2010) and various environmental NGOs
(e.g. Forum Ökologisch-Soziale Marktwirtschaft 2010). Also the German Energy Agency,
DENA, criticised the proposed cuts and demanded to increase the funding to €5 billion per
year i.e. more than ten times the proposed level (ENBAUSa 2010a).
Note that while the shifting of budget from a later year to an earlier year described above
does not qualify as a spending cut when looking at the whole period, the cuts announced
by Ramsauer were actual cuts reducing the budgetary commitment made for the period.
206
Case II: CO2 Buildings Rehabilitation Programme in Germany
More funding for CBRP
Responding to a parliamentary question by the Green Party (Deutscher Bundestag 2010e),
the Government announced that the CPRB would receive an extra €5
year
million for the
11 increasing the funds available from € 36 million to €936 million (Deutscher
Bundestag 2010b). The money was supposed to come out of the Energy and Climate Fund,
a Government owned fund to finance low-carbon technologies such as energy efficiency
and renewable energy. The fund should be financed by the large energy suppliers which
were obliged to pay a windfall tax on profits due to delaying the phase-out of nuclear
power plants and windfalls gained from EU Emissions Trading Scheme permits.
The introduction of the Energy and Climate Fund meant that for the first time in the
history of the KfW loan and grant programmes part of the funding was not based on the
budget alone but also on a separate fund. Precisely for that reason the government argued
that the fund offers more long-term certainty as it is independent of budgetary
considerations that take place year on year and adds more flexibility (Deutscher Bundestag
2011a). Minister Peter Ramsauer stated:
I always stressed that we need continuous and stable funding; because a lot of businesses
specialised in the area of CO2-Building Rehabilitation. One cannot ramp up their activities in
one year and slow them down in the next year. [my translation] (Ramsauer 2011)
This is an interesting statement given the fact that the Minister himself proposed to cut the
CBRP’s budget just a year before.
The announcement of the new finance mechanism was heavily criticised in a common
press statement by a wide range of organisations, including construction industry
associations, tenant organisations, consumer associations, and environmental NGOs. Their
207
Case II: CO2 Buildings Rehabilitation Programme in Germany
main objections related to the overall level of funding, which they deemed too low, and the
uncertainty of the payments through the Energy and Climate Fund (Mieterbund 2010).
Interviewees representing the Green Party shared those concerns (Interview G7, G12) and
a Member of Parliament of the SPD stressed that it is important to keep the funding for the
CBRP under control of the parliament rather than implementing funding streams
independent of the federal budget (Interview G9).
Those concerns should soon be confirmed. As already said, the Fund was supposed to be
financed to a significant extent via a tax on windfall profits from extending the phase-out
of nuclear energy. However, after the nuclear accident at Fukushima the German
government decided to revert from the extension of the phase-out and speed up the phaseout process. Therefore, the windfall tax was made obsolete and could no longer fund the
Energy and Climate Fund (Deutscher Bundestag 2011b). In addition to the funding issues
around the windfall tax on profits from extending the phase-out of nuclear power there
were also problems with the second funding stream, the revenues from auctioning EU ETS
permits. Permit prices fell below €1 per tonne of CO2, way below the anticipated €17
when designing the Energy and Climate Funds (Bloomberg 2012). Ironically, the aim of
providing finance for the CBRP independent of the federal budget in order to reduce the
unstable funding and dependence on budgetary cycles led to even more uncertainty about
the future funding of the programme. This caused controversial political debates with the
opposition demanding to put the CBRP back into the overall budget (Deutscher Bundestag
2012). Paradoxically, even Minister Ramsauer, who previously announced the 50%
spending cut for the CBRP in 2010, urged the Coalition to revoke the cuts and top up the
CBRP in 2012 to make up for the reduced funding coming from the Energy and Climate
Fund (Neuerer 2011). However, Members of Parliament of the CDU and FDP suggested
208
Case II: CO2 Buildings Rehabilitation Programme in Germany
that even if there were shortfalls in the Energy and Climate Fund, the CBRP would not be
affected significantly by any funding cuts (Interview G10, G11). For 2012 this prognosis
proved correct, but only because of a special loan arrangement made to shift funds from
the regular budget to the Energy and Climate Fund.
Summary
Budgetary issues affected the CBRP in a number of ways. First, the shifting of funding
from one year to another resulted in more funds than originally planned in 2009 leaving a
shortage of funds for 2010. While the overall amount of finance was not reduced, it caused
some controversy with some groups labelling the lower funding levels in 2010 as cuts.
Second, the growing public debt eventually led to the introduction of the debt brake which
prohibits the government to take on new net debt. The debt brake was used as an argument
to cut funding for the CBRP by 5 % for
11. Ironically, the expansion of the CBRP’s
funding in 2008 (see section 5.2.3) contributed to public debt and in that sense had
negative feedback effects. Third, the introduction of a budget-independent Energy and
Climate Fund was supposed to provide more long-term certainty and stability of the
finance stream. However, it looks like the introduction of the Fund may have caused the
opposite being based on rather unreliable finance sources itself. Again, even though
unintended, the modified finance structure started to generate negative policy feedback
effects in the form of unreliable funding for the programme.
To conclude, the debate remains unresolved and there is no certainty about the future
funding for CBRP. Many interviewees stressed that the unpredictability of funding had
negative effects on the market in the past (e.g. interview G7, G9, G10, G12), although one
209
Case II: CO2 Buildings Rehabilitation Programme in Germany
interviewee disagreed and stated that the private housing market would not be affected to a
large extent (Interview G1).
5.2.5 Change of personnel
Tied in with other drivers such as budgetary constraints, the change of personnel at the
ministerial level in late 2009 had an impact on the CBRP. Peter Ramsauer (CSU)
succeeded Wolfgang Tiefensee (SPD) as minister of the BMVBS. Of course one would
expect some change in priorities related to the change in government and a new coalition
plus a change of the minister, but Ramsauer’s appointment had an impact beyond party
politics.
Many interviewees described Ramsauer as a ‘transport minister’ rather than a ‘construction
minister’ who had prioritised investment into the transport infrastructure and neglected
policies focusing on building construction and refurbishment such as the CBRP (Interview
G1, G2, G5, G7, G9, G10, G12). A good example to illustrate this was the announcement
of cuts to the CBRP budget in 2010.
CBRP budget cuts
As already mentioned above, Ramsauer announced in June
1 that the CBRP’s budget
would be cut by 50% in 2011 (Ramsauer 2010). Ramsauer’s announcement was followed
by unusually harsh criticism from industry associations, tenant organisations, and NGOs.
The Bundesvereinigung Spitzenverbände der Immobilienwirtschaft (BSI), a federation of
the large real-estate associations, claimed that the cuts were ‘nonsense’ and ‘economically
and fiscally wrong’ (BSI 2010). On a similar note the Bundesverband deutscher
Wohnungs- und Immobilienunternehmen (GDW), a real-estate association, labelled
Ramsauer‘s
decision
a
‘crass
political
210
misjudgement‘
(GdW
2010).
The
Case II: CO2 Buildings Rehabilitation Programme in Germany
Baugewerbeverband Schleswig-Holstein, a regional construction industry association,
even went so far to call the cuts ‘dishonourable’ (BAULINKS 2010). Also the German
Energy Agency (DENA) criticised Ramsauer’s proposals and suggested that rather than
cutting the CBRP’s budget it should be raised to €5 billion per year (BAULINKS 2010).
Also of interest, Ramsauer faced opposition by his own ministry: the parliamentary state
secretary of the BMVBS, who belonged to the FDP, demanded publically the CBRP
funding be increased to a level of €3 billion per year. This was just one day before
Ramsauer defended the cuts in parliamentary debates (Schäfer 2010). Furthermore, just a
few weeks earlier the Environment Minister Norbert Röttgen, also belonging to the
CDU/CSU, criticised the cuts in a leaked confidential paper. In the paper he said:
These cuts will lower the refurbishment rate massively and will have drastic impacts on the
economy and the job market. [my translation] (Rheinische Post 2010)
Also the responsible rapporteur of the CDU/CSU in the parliamentary budget committee
and the spokesperson for construction and transport expressed their regret about the cuts
(Spiegel 2010).
The opposition from within his own ministry and party shows that Ramsauer did not
simply represent an official party or ministry position, but rather followed his own agenda
on the matter. As one interviewee put it:
This is actually unusual, everyone fights for his budget. And if it is about adjustments within a
given budget one could of course take the position ‘okay, he needs to trade the maintenance of
motor ways against the CBRP. Of course we would have wished for a clear position pro CBRP,
but this was not even the main issue. Apparently he did not make the case with Schäuble [the
then-finance minister] for keeping or increasing the budget for the CBRP. Instead, the
Programme’s budget was cut. [my translation] (Interview G2)
211
Case II: CO2 Buildings Rehabilitation Programme in Germany
One should, however, also take into account the on-going debates at the time around
budget constraints and the debt brake (see section 5.2.4). Ramsauer linked his arguments
to this debate and embedded the cuts in a wider discussion about public spending.
Summary
Shortly after being appointed, Peter Ramsauer announced cuts in the CBRP’s budget. It
seems that while impossible to prove, a different minister with more balanced priorities
giving more consideration to the overall coalition parties’ positions would have reacted
differently and tried to prevent the cuts to the CBRP’s budget. Ramsauer resisted pressure
not to cut the funding for some time and his resistance was a source of friction. However,
as sections 5.2.1 and 5.2.4 have shown, following the 2010 Energy Concept and with the
formation of the Energy and Climate Fund, financial resources equivalent to the cuts
initiated by Ramsauer were made available to fund the CBRP.
Conclusion
While the CBRP is a fairly technical and sophisticated policy, its history is marked by
frequent, and often unpredictable, changes as a result of politics. Politics transformed the
CBRP in various ways, at times supporting its expansion but at others constraining the
programme’s effectiveness. Throughout the CBRP, climate change policy has been a
major driver but with varying degrees of importance. Other drivers, for example support
for the construction industry and job creation, sometimes reinforced the pressure from
climate change policy, but there were also incidents where pressures, such as budgetary
constraints or the change of minister, had regressive impacts on the programme. In these
cases, it seems like ‘politics determined policy’.
212
Case II: CO2 Buildings Rehabilitation Programme in Germany
However, there is also evidence for the opposite dynamic whereby ‘policy determined
politics’. For example, during the recession the government used the CBRP as a tool to
stimulate economic activity. But this initiative had important negative policy feedback
effects. The enlargement of the CBRP led to higher public debt, which in itself was used
by the responsible minister as an argument to cut the CBRP’s budget. Hence the initial
expansion of the programme ironically contributed to factors that constrained, and even
reverted, the CBRP’s development. Another of these feedback effects, but in this case a
positive policy feedback effect, is that with growing budgetary size, the CBRP also
attracted strong lobby groups across different sectors including industry associations,
unions, homeowners, tenants, and environmental NGOs making it more and more difficult
for the Government to reduce the funding for the programme.
There are a number of observations that can be made. First, the funding mechanism and
dependence on annual budgets has put significant strain on the programme in recent years
and poses a risk to the long-term stability of the CBRP. Second, in the past short-term
political decisions were accompanied by unintended consequences unfolding over the
medium- and long-term. A good example is the creation of the Energy and Climate Fund,
which was supposed to provide more funding stability and fill the finance gap created by
budget cuts. In fact, the funding base for the CBRP is now even more fragile than it was
before.
If the carbon reductions required for reaching the ambitious climate targets are going to be
achieved with the CBRP, a long-term strategy is needed to provide the resources for the
programme. Otherwise it is likely that, like in the past, political opportunism and shorttermism will prevail.
213
Evaluation of Policies
6 Evaluation
I really reject that kind of comparison that says, Oh, he is the best. This is the second best. There
is no such thing.
Mikhail Baryshnikov (unknown)
This section provides an evaluation of the two policy instruments in terms of their
effectiveness, i.e. the carbon emissions and energy saved compared to the financial
resources spent. Both programmes are applauded by various stakeholders, nationally and
internationally. Due to the CBRP,
ermany is often labelled a ‘front runner’ (Murphy et
al. 2012) in energy efficient building refurbishment. Lowe (2009) highlights that Germany
is one of the few countries in the world that has a large-scale funding programme for
energy efficient refurbishment, and Boonekamp and Eichhammer (2007, p. 273) call the
CBRP a successful example of ‘long-term financial efforts’ with ‘considerable impacts in
terms of energy savings and CO2 emissions reductions’.
Similarly, the Britain’s SO is frequently hailed as a successful policy instrument. The
International Energy Agency (IEA) has ‘strongly commend[ed] the UK government for the
creative approach to energy efficiency taken with the [SO 2002-
8] Commitment’ (IEA
2007) and considers the ‘Energy Efficiency Commitment an impressive success’ (IEA
2006). Similar praise can be found elsewhere (Bertoldi and Rezessy 2008; Forfori 2006;
Lees 2007b)
These commendations are largely based on evaluations carried out for the government
responsible for the funding and regulatory framework.
214
Evaluation of Policies
Based on official statistics, this chapter investigates and compares the SO and CBRP by
exploring: (a) the total quantity of energy and CO2 that appears to have been saved through
the public costs (public costs are defined here as the total contribution made by the public
to fund the policy instrument – in case of the SO this is through energy bills and in case of
the CBRP via taxes); (b) the economic efficiency of energy saved through the public costs,
in Euros per kilowatt-hour (€/kWh); and (c) the economic efficiency of CO2 saved, in
Euros per tonne (€/t). It is beyond the scope of this thesis to carry out a fully fleshed out
evaluation of the two programmes – this would be a herculean task. Instead, it relies on
official figures provided by evaluations commissioned by the respective governments and
their agencies.
However, a substantial body of literature covering several decades of energy research
suggests estimated savings in evaluations are often higher than actual, measured savings
(Clinton et al. 1986; Fels and Keating 1993; Frondel and Schmidt 2005; Gillingham et al.
2006; Hewett et al. 1986; Hirst 1990; Hirst and Hannon 1979; Hirst et al. 1983; Joskow
1993; Nadel and Keating 1991; Newcomb 1984; Vine 1994; Vine and Sathaye 2000).
To make robust judgments of an energy saving programme’s economic effectiveness, we
need to know how much energy and CO2 is actually being saved through the financial
support they provide. But most evaluations of home energy efficiency programmes depend
on calculated, rather than measured, levels of heating energy consumption. This fails to
take into account the discrepancies that have been observed in practice, between calculated
and actual heating energy consumption both before and after refurbishment (Schröder et al.
2011). Factors contributing to such discrepancies include rebound effects, prebound
effects, free rider effects, and technical shortcomings. Those factors will be addressed in
215
Evaluation of Policies
detail in the following and the evaluations of each policy instrument will be critically
analysed regarding their completeness with regard to these reduction factors.
The remainder of this chapter proceeds as follows: Section 2 compares the official results.
Section 3 introduces and offers estimates of the rebound and prebound effects in energyefficiency refurbishment in Germany and the UK, it also discusses free rider effects and
their potential impact. Sections 4 and 5 review existing evaluations of
ermany’s CBRP
and the Britain’s SO programmes respectively, offering a critical analysis before the
chapter concludes.
6.1 Comparison of official results
The accounting methodology for energy and carbon savings differs in the UK and
Germany, hence the following comparison carries inherent limitations.
The German evaluations performed for the CBRP calculate annual reduction of carbon
emissions rather than the induced lifetime CO2 emissions reductions. However, the annual
reductions can be converted into lifetime emissions saved by assuming an average lifetime
of measures of 30 years, as shown by an additional analysis complementing the evaluation
of Clausnitzer et al. (2007) by Gabriel and Balmert (2007).
The figures used in Great Britain for the SO (EEC 2 and CERT) do not include the energy
and carbon savings carried over from previous obligation periods; only those savings
actually achieved under the respective scheme are considered. Savings under EEC 2 are
only reported in TWh, and conversion factors from the Department for Environment, Food
and Rural Affairs (DEFRA) have been used to convert energy savings into carbon savings
taking into account the proportion of different fuels saved as reported by OFGEM. In order
to get annual figures for saved carbon emissions the total carbon savings of EEC 2 were
allocated on a pro-rata basis according to annual energy savings. For CERT saved carbon
216
Evaluation of Policies
emissions have been reported on an annual basis by OFGEM but no figures for energy
savings are available. Using the same energy/carbon ratio as EEC 2, annual energy savings
were calculated for CERT.
The figures are not readily comparable for several reasons: first, the evaluation
methodology applied changed over time (for example, energy savings were discounted
under early versions of the SO, but this is not any longer done). Second, the figures
calculated for this paper are based on the assumptions outlined above and need to be
revised for a like-for-like comparison.
This paper’s detailed comparison of savings and expenditure covers the period 2002-2007
only, as the CBRP started in 2001, detailed evaluations of the SO began in 2002, and the
last evaluation of the SO was published in 2008. SO figures refer to financial years
whereas the CBRP data is based on calendar years.
Carbon and energy savings
From 2002-2007 the CBRP generated savings of about 74 Mt CO2 (lifetime) compared to
savings by the SO of 79 Mt CO2 (lifetime). In more recent years, the SO delivered higher
savings than the CBRP. Figure 1 displays these savings over time.
217
Evaluation of Policies
Figure 17: CO2-emissions in Mt (lifetime) and TWh (lifetime) saved per year 10
Source: own calculations based on various sources (Clausnitzer et al. 2011; Clausnitzer et al. 2010; Clausnitzer et al.
2007, 2008, 2009; Doll et al. 2008; Gabriel and Balmert 2007; IER and PROGNOS 2004; Kleemann and Hansen 2005;
Kleemann et al. 2003; Kuckshinrichs et al. 2010; OFGEM 2005, 2006, 2008c, 2009c, 2010b, 2011b; OFGEM and
Energy Saving Trust 2003; Rosenow 2011); data for energy savings of CBRP based on CO2 savings in 2004 using same
carbon/energy ratio as other years
Concerning energy savings, the picture looks slightly different: the CBRP saved about 173
TWh (lifetime) whereas the SO resulted in 235 TWh (lifetime) from 2002 to 2007, i.e.
35% more than the CBRP. The difference is most likely due to the fact that households in
the UK use a higher proportion of gas for heating (DECC 2011c) which has a lower carbon
factor than heating oils that are used in Germany (EEFA 2010). Also the carbon factor for
electricity is higher in Germany than in the UK (DECC 2012a; UBA 2012). These
circumstances are likely to result in higher energy savings per unit of carbon saved for the
SO. Lifetime energy savings are not discounted.
Financial resources spent
The financial resources for the two instruments are provided via two very different routes:
in case of the SO the energy customers paid for the scheme with their bills, while the
10
SO data is for financial years whereas CBRP data is for calendar years.
218
Evaluation of Policies
CBRP funding was based on taxpayers’ money. In both cases households receiving
financial resources for energy efficiency measures make an additional contribution to the
overall cost. In some cases third parties such as social housing providers and local
authorities may also add funding. However, in order to be able to compare the two
instruments it is reasonable to exclude these contributions and compare only the cost that
the public bears either through tax or a premium on the energy bill.
From
to
7 about €1.5 billion were spent by energy suppliers to comply with the
SO including both direct costs and indirect cost (Lees 2006, 2008). In the same period
federal funding for the CBRP amounted to as much as € .8 billion (BMVBS 2011). Table
10 relates the cost figures to the energy and carbon savings. It should be noted that these
costs include public costs only, not the portion of refurbishment costs paid by the
homeowners.
Table 10: Total expenditure, carbon and energy savings 2002-2007*
Carbon emissions
Energy saved
saved in million t
in TWh
CO2 (lifetime)
(lifetime)
SO
79
CBRP
74
Public costs in
Public costs
Public costs in €
billion €
per t CO2
235
1.5
19
0.6
173
2.8
38
1.4
€ cent per
kWh
Source: own calculations based on various sources (BMVBS 2011; Clausnitzer et al. 2011; Clausnitzer et al. 2010;
Clausnitzer et al. 2007, 2008, 2009; DECC 2010b; Doll et al. 2008; Gabriel and Balmert 2007; IER and PROGNOS
2004; Kleemann and Hansen 2005; Kleemann et al. 2003; Kuckshinrichs et al. 2010; Lees 2006, 2008; OFGEM 2005,
2006, 2008c, 2009c, 2010b, 2011b; OFGEM and Energy Saving Trust 2003; Rosenow 2011)
* figures for SO from April 2002 to March 2008, figures for CBRP from January 2002 to December 2007; used factor of
1.16 ( 1/ 7/ 13) to convert £ in €
The table shows that, based on the official results, the CBRP requires 2 times the financial
resources per t CO2 (lifetime) or TWh (lifetime) as the SO. One reason could be that
erman dwellings are more energy efficient than the UK’s (BPIE 2011), which can
therefore benefit proportionately more from cheaper upgrade measures at lower levels.
Further, CBRP subsidies are only given for top-end retrofits, where the marginal return per
219
Evaluation of Policies
Euro invested is low (Galvin 2010). Note that the cost figures do not include the benefits in
terms of saved energy costs. For a comparison to other carbon reduction programmes those
would need to be included. But even without considering the energy savings, the costs per
t CO2 are still below other well-known alternatives such as on-shore wind power
(~6 €/tCO2) or solar power (>5
€/tCO2) (McKinsey, 2007) .
The domestic price for electricity in both countries is well above 15 Cent/kWh, for gas and
heating oil this is more than 6 Cent/kWh, which means that the public cost for the two
programmes is more than recovered by the energy savings.
6.2 Evaluating energy efficiency programmes
The figures presented above rely on the evaluations commissioned by the respective
governments and agencies. I now discuss which factors should be taken into account in
evaluations of energy efficiency programmes followed by a critical analysis of the
methodologies used in the evaluations of the CBRP and the SO.
6.2.1 Rebound effects
Jevons (1865) argued that 'it is a confusion of ideas to suppose that the economical use of
fuel is equivalent to diminished consumption. The very contrary is the truth.’ It is argued
that increasing energy efficiency makes it cheaper to produce goods and services, leading
to greater wealth, which drives energy consumption higher than before the efficiency
improvements. This hypothesis has also been labelled back-fire (Brookes 1990).
In reality, the evidence available shows that only a proportion of the energy savings is
taken back in the form of increased usage of energy services. This phenomenon is known
under the term rebound effect and occurs on the micro level, both directly and indirectly
(Greening et al. 2000; Sorrell 2007), and also on the economy-wide, macro level (Barker et
al. 2007; Sorrell 2007). The present study is concerned only with direct micro rebound
220
Evaluation of Policies
effects, namely the percentage of the energy saved through refurbishment that is taken
back, to provide increased thermal comfort or convenience after refurbishing.11
There are various assessments of the magnitude of the rebound effect for domestic heating.
Greening et al. (2000) reviewed 75 studies and found rebound effects of 10-30% for space
heating. A review of over 500 studies for the UK Energy Research Centre found rebound
effects for home heating averaging less than 30% (Sorrell 2007). Haas and Biermayr
(2000) estimate a figure of 30% for home heating in Austria, a country with similar
climate, building standards and indoor culture to those of Germany.
Localised studies in Germany show similar trends. Erhon (2007) found actual consumption
increasing steadily above calculated values as thermal standards became higher, i.e. as
calculated consumption, given here in kilowatt hours per square metre of floor area per
year (kWh/m2) became lower. This started at 0% for detached houses of 150 kWh/m2, and
multi-dwelling buildings of 100 kWh/m2, and reached 50% for the most energy-efficient
dwellings of both types. Loga et al. (2011) found similar effects in homes with calculated
consumption ratings lower than 50 kWh/m2, while Kaßner et al. (2010) found actual
consumption of 65% above calculated for dwellings with calculated ratings of 75 kWh/m2.
Thomsen et al. (2005) reported 100% over-consumption (i.e. above the expected after
refurbishment) in a study of ‘advanced solar low-energy buildings’. A recent study
suggests that the rebound for space heating ranges between 12% for owners to 49% for
low-income tenants (Madlener and Hauertmann 2011).
reller et al.’s (2010) study of
100,000 dwellings, referred to above, revealed progressive magnitudes of over-
11
Short of a detailed thermodynamic examination of the building envelope and heating system it is
impossible to know what portion of this difference is genuine ‘take-back’ and what is due to technical
shortcomings in the refurbishments.
221
Evaluation of Policies
consumption increases in parallel with the progressive tightening of legal standards over
the past decade. These figures are not strictly the ‘rebound effect’ because their baseline is
current calculated consumption, rather than calculated savings in consumption. But they
are useful measures as they show the same phenomenon: over-consumption compared to
the calculated rating, in low-energy dwellings.
The picture in the UK is similar. A study by Milne and Boardman (2000) suggested that
rebound effects for insulation measures are likely to be around 30% in the UK. For
electrically heated dwellings Henderson et al. (2003) reported rebound effects of 18%.
Without quantifying the contribution of the rebound effect, the total measured shortfall
from expected savings in gas heated homes is 55% (Energy Saving Trust 2004). This
figure includes technical shortcomings and other reduction effects which are not rebound
effects. The rebound effects for fuel-poor households in the UK may be as high as 65100% (Hong et al. 2006). However, Martin and Watson (2006) showed that although lowincome customers displayed slightly higher rebound effects, there was little difference in
rebound effects between average and low-income customers. More recent analysis of
seven popular energy efficiency measures arrived at much lower figures of 5-15% for
direct and indirect rebound effects (Chitnis et al. 2012).
6.2.2 Prebound effects
Sunikka-Blank and Galvin (2012) reviewed eight studies which analysed measured and
calculated consumption figures for a total of 3400 German dwellings, plus measured
ratings only for an additional 1 million. They found a clear trend was evident in the
studies: on average, the higher the calculated energy rating (i.e. the less energy-efficient
the building), the larger the percentage difference between measured and calculated
consumption. The authors labeled this the ‘prebound effect’, as it is the opposite of the
222
Evaluation of Policies
rebound effect. It refers to the percentage by which the measured consumption of a
dwelling is lower than the calculated consumption. Note that this is not the result of an
intervention (as it is in case of the rebound effect), but the result of the way actual
measured consumption differs from the modeled consumption. It is an inaccuracy in the
models used rather than a result of energy efficiency refurbishments.
For example, a dwelling with a calculated rating of 200 kWh/m2 that is consuming 160
kWh/m2 is showing a prebound effect of 20%. Though there were slight differences in the
results of the studies reviewed, generally the prebound effect was, on average, 0% for
dwellings of calculated rating 100 kWh/m2, rising through 30% for ratings of 220 kWh/m2,
up to 55% for ratings of 500 kWh/m2. While these were average values, the measured
consumption of individual dwellings varied widely (Greller et al. 2010), Bild 1, for a
graphical representation of the range of differences). However, average figures are useful
for the present study, because, in the absence of evidence to the contrary, I may presume
that the discrepancies between calculated and measured heating energy consumption in
buildings that are about to be refurbished belong to the same statistical distribution as
those in the total residential building stock.
Sunikka-Blank and Galvin (2012) compared these findings to those of comparable studies
in France (Cayre et al. 2011), the Netherlands (Tighelaar and Menkveld 2011), Belgium
(Hens et al. 2010) and the UK (Kelly 2011) and found similar trends.
6.2.3 Free rider effects
A point of controversy in the energy efficiency literature has been the issue of free riders
and their impact on the effectiveness of programmes (Blumstein 2010; Eto et al. 1995;
Gillingham et al. 2006; Joskow 1993; Kreitler 1991; Levine et al. 1995; Malm 1996; Train
1994; Vine et al. 2012; Vine 1994; Vine and Sathaye 2000; Wirl 2000). Free riders ‘are
223
Evaluation of Policies
programme participants who would have purchased and installed an energy efficiency
measure even in the absence of the programme’ (Geller and Attali 2005, p. 18). This
phenomenon is also referred to as ‘deadweight’ (Lees 2007b).
Kreitler (1991) estimates that free rider effects can reduce estimated savings by up to 80%,
and Loughran and Kulick (2004) estimate this at 50-90%. Malm (1996) estimates free
rider effects on high efficiency heating system purchases at almost 90%. A crossevaluation survey on US demand side management programmes resulted in estimates of 050% with an average of 12.2% (Eto et al. 1995).
A recent study in Germany (Grösche and Vance 2009) calculates that up to 50% of
estimated savings may be lost due to free riders. The impact of free riders depends very
much on the type of measure and the specific context in which the measure is provided.
Responding to the critics, some stress that while free rider effects exist, they are offset by
so-called ‘free driver’ or spillover effects, which result from measures being installed as a
result of but not through the programme. This is the case when people install additional
measures over and above the programme’s incentives or if non-participants take up
measures as a result of the programme (Eto et al. 1995). Recent evaluations of energy
efficiency programmes show that spillover effects can be substantial and counterbalance a
significant proportion of the free rider effects (NYSERDA 2012).
6.3 Evaluations of the Supplier Obligation
DECC commissioned independent evaluations of EEC 1 and EEC 2 conducted by Eoin
Lees Energy (Lees 2006, 2008). In addition, OFGEM provided annual reports on the
results of the SO since 2002 (OFGEM 2003, 2004b, 2005, 2006, 2007, 2008c, 2009c,
2010b, 2011b) and, with support of the Energy Saving Trust, a summary review of EESoP
1-3 covering 1994-2002 (OFGEM and Energy Saving Trust 2003). Those reports result
224
Evaluation of Policies
from the regulator’s duty to report to the Secretary of State on the programme. Until the
end of EEC 2, OFGEM reported the accounted lifetime savings in terms of TWh broken
down by energy supplier and measure. Since 2008 the target metric changed to t CO 2
(lifetime) and energy savings are no longer reported.
This paper focuses on the independent evaluations carried out by Lees (2006, 2008)
because the OFGEM reports are more an accounting procedure based on an ex-ante agreed
scoring system than a comprehensive evaluation. Lees critically discusses assumptions
made in the accounting process and contrasts those with ‘real’ data. The evaluations
include detailed estimates for energy savings broken down by measure and the cost of the
SO to energy suppliers, receiving households, and others.
6.3.1 In-depth appraisal of the evaluation
The evaluations carried out by Lees (2006, 2008) essentially correct the energy saving
scores used by OFGEM to account for the savings achieved by the obligated parties.
Because the SO always focused on small energy users – first households and small
businesses and then just households – the approach taken by the regulator to verify the
savings was based on deemed savings rather than measured savings. The values for those
deemed savings were generated by using models such as the BRE Domestic Energy Model
(BREDEM). Measures for which reliable engineering estimates of energy savings existed
were given a specific energy savings score (later carbon emissions savings).
According to Lees (2008), the following steps were carried out to correct the energy
savings reported by OFGEM in the evaluations:

remove any uplift factors from the energy saving values: some measures (called
innovative action) received higher energy saving scores in order to incentivise
225
Evaluation of Policies
additional technologies with a lower cost effectiveness. These uplifts are subtracted
in the evaluations.

correct for the heat replacement effect for lighting and appliances which results in
increased fossil fuel usage: more efficient forms of lighting and appliances result in
less waste heat which is subsequently compensated by increased space heating.

remove the fraction of energy savings taken in the form of increased comfort: see
section 6.3.2 on rebound effects.

update the energy saving values for insulation measures and the lifetimes for
heating measures to be consistent with those used in the most recent obligation
period. This is based on the assumption that most recent scores have been revised
based on new evidence.

convert from fuel standardised units to the actual units of the fuel: EEC 1 and EEC
2 used a fuel-weighted accounting system depending on the carbon content of the
fuel saved.
The SO evaluations also account for free rider effects (section 6.3.3). Prebound effects are
addressed by adjusting the modelled savings based on measured savings across a sample of
households (see section 6.3.3)
6.3.2 Prebound and rebound effects
Rebound effects are partially taken into account in evaluations of the SO. However, this is
related to space heating and insulation only.
During EESoP 1 the energy savings scores used were based on BRE advice and the
BREDEM model and modified with EST assumptions on comfort taking (rebound effect).
BRE had carried out research during the Energy Efficiency Demonstration Scheme and
also the Energy Efficiency Best Practice Programme which followed it on energy savings
226
Evaluation of Policies
arising in case studies and also paired test houses. It is not known how the figures BRE
produced was weighted between the two routes (model and test houses), but I was told that
it is likely to have been heavily influenced by the paired test houses (Eoin Lees, personal
communication 29/06/2013).
Subsequently, comfort factors, energy saving values changed slightly due to improving
average efficiency of the boiler stock over the various phases. Only later were ‘technical
reduction factors’ introduced which then became ‘in use reduction factors’ factors in the
successor of CERT, the Energy Company Obligation (ECO). Because energy saving
scores were not only based on modelled energy savings but on actual savings in test houses
it is likely that the prebound effect was at least addressed partly by the methodology
chosen.
When attributing saving scores during EESoP 1 it was assumed that households would
take 30% of potential savings from insulation measures in improved comfort. For EESoP
2 and 3 it was assumed that on average households would take 42.5% of savings as
increased comfort. For non-building fabric measures, only the fitting of energy-efficient
lighting (CFLs) was considered to have a small rebound effect of 2.5%.
The EEC 1 evaluation used a 30% rebound factor for the calculation of savings from
insulation (Lees 2006), the same factor was used when setting the energy savings score for
the different measures for EEC 2. For low-income households it was assumed to be 45%
with 15% for able-to-pay households.
However, the comfort factor was reduced to 15% during CERT (Lees 2008) as a result of
research commissioned by DECC on discrepancies between actual and estimated savings
(Sanders and Phillipson 2006). DECC states that analysis allows for a comfort factor of
40% for insulation measures in the super priority group, a group of customers thought to
227
Evaluation of Policies
be most vulnerable (15% in priority group, a group of those over 70 and on certain
benefits) and 25% for heating measures in the super priority group (0% in the priority
group) (DECC 2010a). It is likely that similar figures will be used in the CERT evaluation.
Similarly to Germany, the modelled heating energy consumption pre-refurbishment is
higher than actual consumption. However, the main difference to the evaluations of the
CBRP is that since the beginning of the SO those effects have effectively been cancelled
out already at policy design stage by frequent revisions of the energy savings scores used
in the target setting and accounting process that follows. The revisions are based on
observed savings within a sample of recipients of measures rather than modelled
hypothetical savings. Research by the Building Research Establishment (BRE) in the
198 s showed the gap between the two and suggested an overall ‘reduction factor’ of 3 %
(Henderson et al. 2003), and there is evidence that it still exists (Kelly et al. 2012). This
reduction factor includes rebound effects, prebound effects, and technological
shortcomings of measures. Although the prebound effect is not explicitly quantified for
every measure, it is included in the part of the reduction factor applied to energy savings
classified as ‘other’.
Further verification was undertaken by assessing the validity of the energy savings scores
used for attributing savings to the various measures. A range of studies was commissioned
by government and stakeholders (AEA 2009; AEAT 2004; BRE 2006; Energy Monitoring
Company 2008; Energy Saving Trust 2004; Henderson et al. 2003; Hong et al. 2006;
Martin and Watson 2006; Sanders and Phillipson 2006). The latest research on energy
savings is currently conducted as part of the National Energy Efficiency Data framework
(DECC 2011e, 2012e). The results of the above studies were used to optimise the deemed
savings scores for future obligations.
228
Evaluation of Policies
Table 11 summarises the energy saving scores used, the lifetime assumed and the comfort
factor applied for one example measure, cavity wall insulation, which was one of the key
energy efficiency measures delivered by the SO.
Table 11: Reduction factors applied to cavity wall insulation in various periods of the SO
SO Phase
Annual energy
savings
(kWh/year)
Period savings
counted over
(years)
Comfort/in use factor
then applied
Comments
EESoP 1
3278
15
30% across all incomes
Off peak electricity
savings only.
EESoP 2
Not identified –
probably same as
EESoP 1
15
42.5% (averaged with
higher for low income)
Electricity
savings
mainly but gas if part of
block.
EESoP 3
Decreased
by
6.8%
from
EESoP 2 (revised
version
of
BREDEM)
40
42.5% (averaged with
higher for low income
Electricity and gas now
obligated – savings
values depend on.
EEC 1
5470
40
45% PG comfort and
15% NPG comfort
factors used
Energy savings values
and
target
include
comfort and average
over all fuels.
EEC 2
Not identified –
probably same as
EESoP 1
40
as EEC 1
as EEC 1
Source: Eoin Lees, personal communication 29/06/2013
It is beyond the scope of this paper to validate the reduction factors applied for the various
measures, but in theory, this process accounts for both rebound and prebound effects at the
policy design and evaluation stage.
6.3.3 Free rider effects
In Britain, free rider effects are explicitly considered in the design of the SO when the
savings scores are defined for different types of measures (Lees 2006, 2008). During EEC
1, free rider effects have been estimated to be less than 30% for all measures other than
DIY loft insulation. The initial factors used for free rider effects were higher than
229
Evaluation of Policies
calculated by the evaluation and total free rider effects were estimated to be 20.7%
compared to the initial assumption of 27.6%. These estimates were taken into account in
subsequent policy design and the correction factors for free rider effects were amended.
The EEC 2 evaluation stresses that free rider effects are below 20% for most of the
important measures (in terms of their contribution to the overall target) except for DIY loft
insulation. Overall, 14% of the savings were attributed to free riders. However, given the
higher free rider effects factors for wet and cold appliances, the overall losses due to free
riders accounted for about 20% (Lees 2008). When the SO was extended in April 2011,
DECC decided that for the remainder of the obligation period there would be negligible
free rider effects being subsidised by the programme. This is due to the high number of
filled cavity walls and insulated lofts (DECC 2010b). However, a detailed analysis is
missing. Previous evaluations systematically assessed free rider effects and it is likely that
the post-CERT evaluation will provide more clarity.
6.4 Evaluations of the CO2 Building Rehabilitation Programme
The first detailed evaluation of the CBRP, Kleemann et al. (2003), analysed the results
achieved in 2001 using the IKARUS space heating model. This gives energy and carbon
savings based on calculated building performance. The figures from the sample were
extrapolated to all the buildings that received KfW funding. No detailed evaluation exists
for 2002-2004, but Doll et al. (2008) provide estimates for the carbon savings achieved in
these years, quoting an unpublished presentation of the Federal Ministry of Transport,
Building and Urban Development (BMVBS – Bundesministerium für Verkehr, Bau und
Stadtentwicklung) from 2006. IER and PROGNOS (2004) also offer figures for energy
savings achieved in 2002 and 2003. In reports prepared for the KfW for 2005-2010,
Clausnitzer et al. (2007, 2008, 2009) and (Clausnitzer et al. 2010) offered detailed
evaluations of the programme, including energy and carbon emissions saved. Diefenbach
230
Evaluation of Policies
et al. (2011) use a similar method in their evaluations for 2005-2010, also prepared for the
KfW.
6.4.1 In-depth appraisal of the evaluations
Clausnitzer et al. (2008) follow the same methodology and format as the bulk of the post2005 evaluations, and may be used as an example. It estimates CO2 and energy savings
achieved through the support of CBRP funding provided in 2007. The owners of 1022
properties that had received CBRP support were surveyed and 658 responded.
Respondents gave details on the type, location and orientation of the building, plus its
thermal features (insulation, windows, heating system, etc.) before and after refurbishment,
and the energy source used for heating. Using these parameters the authors employed a
simplified method, developed by Loga et al. (2005), to calculate heating energy
consumption, in line with energy ratings given in German energy performance certificates.
This gives theoretical results based on the presumed technical quality of the building, and
does not take rebound and prebound effects into account. The authors modified this
according to a building typology classification method (apparently produced by the Institut
Wohnen und Umwelt, but mis-referenced in Clausnitzer and colleagues’ studies) and the
likely effects of each building’s geographical location and orientation to the sun. A
software tool was then used to give the expected energy and CO2 emission savings per
year from this data.
Further consideration was given to the likely technical lifetime of the thermal
refurbishment measures, e.g. 40 years for roof insulation; 20 for boilers (Clausnitzer et al.
2008, p. 54). From these figures the annual energy and CO2 emission savings, produced as
a result of the refurbishments, were calculated. As a cross-check on survey data a ‘subsample’ (number and selection method not given) of properties were visited. The results
231
Evaluation of Policies
were then extrapolated to all domestic properties that received CBRP funding, to produce
national totals. The study concludes that the CBRP in 2007 supported a saving of 330,000
t of CO2 per year and 0.94 TWh of energy per year. Assuming a 30-year average for the
technical lifetime of the refurbishments (Clausnitzer et al. 2008, p. 68), this equates to total
savings of 9.9 million t of CO2 and 28.2 TWh.
From these results one can calculate what this implies for the average dwelling. According
to the study, the subsidies enabled 88,590 dwellings, of total floor area 7.75 million m2, to
save a total of 940 GWh of heating energy per year. This equates to a saving of 121
kWh/m2, or 54% of an average pre-refurbishment consumption of 226 kWh/m2.
The results of Clausnitzer et al. (2010; 2007, 2008, 2009), supported by Diefenbach et al.
(2011) show a sharp fall in the savings per m2 of floor area as from 2008: energy savings
rose steadily, from 110 kWh/m2 in 2005 to 131 kWh/m2 in 2008, then fell to 84 kWh/m2
and 82 kWh/m2 in 2009 and 2010, respectively. This is most likely because prior to April
2009 only whole-house refurbishments were eligible for subsidies, but from April 2009
single refurbishment measures – such as replacing a window or insulating a roof – became
eligible.
6.4.2 Prebound and rebound effects
The prebound effect depends, as noted in Section 2, on the (calculated) pre-refurbishment
heating energy consumption.
Despite the length of the evaluations (over 250 pages), Clausnitzer et al. give only minimal
information on methodology. While their authors must have estimated the (calculated) preand post-refurbishment heating energy consumption, they give only the difference between
these two. Further, their questionnaire asked only about building characteristics, not about
energy bills, so an opportunity to explore the validity of their calculated data was missed.
232
Evaluation of Policies
In the absence of further information, and in line with the growing evidence of prebound
and rebound effects, I estimate the actual savings to be lower than the official figures
quoted in Table 10. Assuming an average 30% prebound effect (all energy) and a 10-30%
rebound effect (heating energy) the total savings from 2002-2007 reduce from 173 TWh
(lifetime) to 84-108 TWh (lifetime) or 74 tCO2 to 36-47 t CO2 at a price of €6 -77/t CO2
and 2.6-3.4 Cent/kWh. As indicated above, energy prices are well above this range and
this makes the policy still cost effective even if rebound and prebound effects are
subtracted. To reiterate, the cost figures do not include either the private cost of the
investment or the value of the energy savings, only the public cost.
A much more sophisticated analysis would be required in order to establish more reliable
figures for carbon and energy savings. Because the rebound and prebound effect differs
depending on the energy use of the building before and after a retrofit such an analysis
would need to be based on disaggregated data breaking down the refurbished houses into
different energy use fragments. Further research could provide insights into the precise
scale of the prebound and rebound effects applying to the CBRP. The main conclusion
from this initial analysis is, however, that the evaluations systematically exclude prebound
and rebound effects which is likely to result in an over-estimate of carbon and energy
savings.
6.4.3 Free rider effects
None of the evaluation studies carried out for KfW (Clausnitzer et al. 2010; Clausnitzer et
al. 2007, 2008, 2009; Kleemann et al. 2003) mention free rider effects. An evaluation of
the CBRP and other policies focusing on home energy efficiency carried out for the
Federal Office for Building and Regional Planning (BBR) stresses that there is no reliable
data for free rider effects and models the implications of 0-30% free rider effects
233
Evaluation of Policies
(Kleemann and Hansen 2005). Only one evaluation study of carbon reduction policies
carried out for the Federal Ministry for Economy and Labour explicitly discusses free rider
effects with regard to the CBRP (IER and PROGNOS 2004). Without quantifying the
effect, the authors of the study provide qualitative estimates for various policy instruments.
For the CBRP they estimate that free rider effects are high. The study highlights that no
reliable data exists and that future research needs to address this gap.
Since the IER study no reliable estimates have been made available (Climate Policy
Initiative 2011; Frondel 2008; RWI and forsa 2008). There is, however, some evidence
that indicates free rider effects. A survey covering 244 households that received energy
efficiency advice shows that 59% of those who invested in energy efficiency measures did
not apply for financial support such as the CBRP (Friedrich 2006). Of the 41% who
applied for financial support for the investment in energy efficiency measures 81%
submitted applications to KfW programmes, most probably to the CBRP. Only 9% of the
41% who did apply for financial support would not have undertaken the investment i.e.
just 4% of all households that did invest in energy efficiency improvements would not
have refurbished their homes without the financial support offered by programmes such as
the CBRP. Of those who applied for financial support 64% stated that they would have
done less energy efficiency improvement without any support. More importantly, 11% of
those who received or applied for financial support stated that they would have refurbished
their homes to the same standard in the absence of any support (Figure 18). This suggests
free rider effects of at least 11% plus an unknown percentage included in the 64% of
households who would have undertaken some but not all of the investment without
financial support.
234
Evaluation of Policies
Figure 18: Financial support for refurbishment and impact on decisions of households
Source: based on Friedrich (2006)
The figures suggest that some of the energy efficiency refurbishments that are supported
by the CBRP would have happened anyway. Given that the study was based on a small
number of households one needs to be cautious when interpreting the results, even though
the overall validity of the study design was critically peer-reviewed and generally
approved by the Wuppertal Institute (Sch le and Hanke
6). In addition to the small
sample size, the study is also more than 6 years old and the CBRP’s requirements did
change over the years. Also, the method used suffers from two problems: a) it may provide
unreliable estimates when the wording of the questionnaire is inappropriate and b) it does
not allow for an estimate of the level of inaccuracy (Vine and Sathaye 2000). Others
question the validity of such methods because intentions are not very good predictors of
actual behaviour. Asking people about their decision and whether they would have done it
235
Evaluation of Policies
without financial support does not produce reliable estimates of free riders (Peters and
McRae 2008).
Still, the evidence illustrates that there is a real need for more research in Germany making
use of a much bigger sample size and generating more reliable data. So far free rider
effects have been either ignored or downplayed in the evaluations of the CBRP. An
evaluation study on the economic effects of the CBRP funded by the KfW argued that an
assessment of whether investments in energy efficiency improvements would have been
undertaken regardless of the financial support provided by the CBRP was outside of the
analysis. The CBRP would raise awareness in the first place and therefore trigger
investments in energy efficiency improvements (free drivers, see section 6.2.3), mainly
through giving advice to households (Kuckshinrichs et al. 2009). However, research on
German energy advice programmes also indicates significant free rider effects of 66-89%
depending on the type of energy efficiency measure (Frondel 2008).
Conclusions
The literature on the evaluation of energy efficiency programmes is heavily dominated by
US scholars, while contributions from German and British scholars are rare. As a result of
critical research, evaluations of energy efficiency programmes in the US became much
more sophisticated over time (Vine 1994). Similar attempts could help improve the
validity of the evaluations in Germany and the Great Britain. This paper is a first step with
its critical evaluation of the evaluations of two prominent energy efficiency policies in
Germany and Great Britain.
The UK evaluations account for all three of the savings reduction effects. Rebound effects
are explicitly subtracted from calculated savings for some measures both in the evaluations
and at policy design stage. Prebound effects are addressed by using adjusted savings scores
236
Evaluation of Policies
based on observed rather than modelled savings (although their contribution to the total
reduction factor applied is unknown). Free rider effects are explicitly considered in savings
estimated from various refurbishment designs. Nevertheless it is beyond the scope of this
paper to assess the accuracy of the estimates of these three factors in the UK.
In contrast, evaluations of the German CBRP do not account for any of these reduction
effects. My critical review of the evaluation for 2007 Clausnitzer et al. (Clausnitzer et al.
2008) suggest that when rebound and prebound effects are taken into account, actual
savings may be less than half as great as the evaluation estimates, quite apart from any
possible free-rider effects. However, the costs per unit of energy saved are still well below
the cost of energy making it cost-effective.
There are increasing calls for actual, rather than estimated, savings to be used in
assessments of energy saved through refurbishments (e.g. Greller et al. 2010; Schröder et
al. 2011).
Despite repeated demands for doing so, in Germany free rider effects are not included in
any of the evaluations of the CBRP. There is insufficient data on free rider effects in
Germany but the evidence that exists suggests that those effects may be significant. A
thorough assessment of potential free rider effects and their systematic considerations in
the evaluations would improve the reliability of the results.
237
Cross-case analysis
7 Cross-case analysis: politics of change
Comparison is the death of joy.
Mark Twain (unknown)
Mark Twain, quoted above, suggests that comparison is not a joyful exercise. I would
argue that while it may not always be joyful, it is often useful. It is useful because such a
comparison, in this case a cross-case analysis, brings both similarities and differences to
the fore.
Following the stages of Systematic Process Analysis outlined in chapters 1 and 3, the
challenge of this chapter is to draw out the identified mechanisms of policy change and
discuss their relation to the theoretical framework and hypotheses developed in chapter 2.
The chapter proceeds as follows: first, the case study material is analysed with regard to
the causal mechanisms of policy change encompassed in the conceptual framework of this
thesis. For each mechanism, its explanatory power is discussed with regard to the case
studies. In some instances, additional mechanisms are suggested, not previously covered
by the conceptual framework. Second, the hypotheses presented in chapter 2 are evaluated
based on the observed changes and the causal mechanisms identified. Finally, the chapter
concludes with discussing the limitations of the analysis.
7.1 Mechanisms of policy change
The conceptual framework contained a number of mechanisms of policy change:

external pressures;

friction; and
238
Cross-case analysis

policy feedback including positive and negative policy feedback.
This section evaluates their explanatory power for both cases in this order.
7.1.1 External pressures
In chapter 2 external pressures have been conceptualised as drivers from a wide range of
possible sources that put pressure on the policy subsystem to change. The conceptual
framework distinguished between short-term events referred to as systemic perturbations
and long-term pressures called subsystem spillovers (Williams 2009). In the following both
cases are analysed with regard to the existence of external pressures and their nature. There
is no clear definition at which point external pressures constitute systemic perturbations or
subsystem spillovers. In this thesis, events such as non-linear price changes over a few
weeks or months are treated as systemic perturbations whereas processes taking years to
unfold are considered subsystem spillovers.
Subsystem spillovers
In both cases the issue of climate change policy constituted an external pressure falling
into the category of subsystem spillovers due to the gradual and long-term nature of the
driver.
While climate change policy affects a whole range of different sectors, it is particularly
relevant for home energy efficiency policy, because homes in the UK and Germany are
responsible for a large proportion of total carbon emissions. In both countries climate
change was a major driver of policy change of the principal home energy efficiency policy.
At its inception, the German CBRP was part of the newly established National Climate
Protection Program, and the purpose of the scheme was also mirrored in its name: CO2
Building Rehabilitation Program (BMU 2000). Extensions of the CPRB in 2003 and 2005
239
Cross-case analysis
were justified with regard to its importance for climate policy. The CBRP also featured in
the 2005 National Climate Protection Programme (BMU 2005), as well as the 2007
Integrated Energy and Climate Programme (BMWi and BMU 2007). Most recently, the
CBRP was highlighted in the 2010 Energy Concept as one of the key policies to reduce
carbon emissions in the domestic sector.
Similarly, the first SO, EESoP1, was partly introduced as a result of national climate
policy: together with the E-factor (the energy efficiency price premium for gas, see below),
EESoP 1 was supposed to raise money for the Energy Saving Trust (EST) that was
established by Government, British Gas and public electricity supply companies in 1992 to
reduce home all energy use and the associated carbon emissions. The EST played a key
role in the
overnment’s climate policy strategy for the domestic sector as outlined in the
1994 UK Climate Change Programme (HM Government 1994). Climate change continued
to play a major role for the SO. Another example is the 2006 Climate Change Bill that
introduced the Carbon Emissions Reduction Target (CERT). While earlier SO scheme
names always referred to energy efficiency, CERT explicitly focused on carbon emissions.
CERT became the most important climate policy in the domestic sector and is also
supposed to deliver the majority of domestic carbon savings in the future (DECC 2009c).
In Great Britain, rising energy prices created another form of external pressure on the SO.
As illustrated in section 4.2.4, energy prices increased steadily after 2003. The rising
energy prices particularly affected the SO in 2008, when prices increased more rapidly
than any increase since the 1973 oil crisis. However, even though there were comparable
price hikes in 2006/2007 there was barely any public debate about energy prices and it had
no effect on the SO. The main reason for the effect of energy prices on the policy
instrument is probably the intervention of macropolitics at this point in time. Interference
240
Cross-case analysis
of macropolitics in subsystem politics is one of the main causal mechanisms leading to
policy change hypothesised by PET (Baumgartner and Jones 1993; True 2000; True et al.
2007a; Walgrave and Varone 2008). In 2008, the involvement of the Prime Minister
himself led to significant and rapid changes in the subsystem of domestic energy efficiency
policy. Attempts to introduce a windfall tax on energy companies’ profits in
8 as a
response to rising energy prices failed due to resistance from within government (more
precisely the Treasury and BERR) and the companies themselves. However, the pressure
to respond politically to rising energy prices and increasing profits of energy companies
accumulated to the extent that eventually Prime Minister Gordon Brown dealt with the
issue directly. Instead of a windfall tax, the government decided to raise the CERT target
and introduce CESP. Without the Prime Minister’s intervention the
% uplift of CERT
would have been rather unlikely, given that CERT had run for just about five months and a
change to the target half way within a SO scheme had never happened before. While one
could argue that rising energy prices fall into the category of systemic perturbation, the
phase of increased public attention to the issue was preceded by a gradual upwards trend in
energy prices. It is more likely that in 2008/2009 a tipping point was reached after pressure
accumulated over a long time period.
An additional external pressure specific to the UK manifested itself in increasing fuel
poverty. Even though it took a long time, fuel poverty is now a well-established policy
driver in the UK (Boardman 1991a, 2010). As illustrated in section 4.2.6, fuel poverty
increased mainly as a result of rising energy prices after 2003. Despite the fact that both
fuel poverty groups and the energy suppliers did not favour using the SO as a policy to
address fuel poverty, although for very different reasons, the SO continued to contain
provisions for targeting fuel poor households (note that it is contested to what extent the
SO actually achieved that (Boardman 2010)). The main reason for such an approach was
241
Cross-case analysis
the lack of any other significant enough policy and the hesitation to establish a fuel poverty
programme based on public expenditure. However, fuel poverty was not always a driver
for the expansion of the SO, the perceived regressive nature of the policy instrument
(because it is paid by energy bills) also constituted negative policy feedback with demands
for scaling it down or replacing it (see section 7.1.3).
Unlike the UK, there is no systematic fuel poverty policy in Germany. The debate has not
gone beyond various pilot projects and statements by different stakeholders. Hence there
are very few references to fuel poverty in the development of the CBRP. It is also notable
that the CBRP is essentially a loan scheme offering loans to those who can prove their
financial credibility. The fuel poor are unlikely to receive such loans (at least not directly,
they may still benefit as tenants of a refurbished property for example)). There is no
convincing explanation for the different level of attention fuel poverty receives in the two
countries. One could argue that fuel poverty is simply not a problem in Germany, which is
why it does not constitute a major policy issue. However, there is evidence that this is not
the case, and that a substantial number of households is affected by it. Unfortunately, there
are no reliable estimates that could be used for a meaningful comparison of levels of fuel
poverty (Kopatz et al. 2010). More research is required to investigate why this issue
became a major policy area in the UK but not in Germany.
Systemic perturbations
A systemic perturbation occurred in Germany in 2008: The global financial crisis and the
following recession had a substantial impact on the CBRP and its financial resources.
Following the financial crisis the German government used the CBRP as part of its
attempts to stimulate economic growth. Although the CBRP’s contribution to economic
prosperity was stressed also prior to 2008, it was only then that the Programme was
242
Cross-case analysis
amended explicitly with the intention of triggering private investment to help overcome
the recession. There were no references to other pressures such as climate change (see
section 5.2.3), indicating that the recession constituted the primary driver at the time.
No systemic perturbations could be identified in the case of the SO, although the issue of
rising energy prices could perhaps be viewed through the lens of systemic perturbations
given the relatively short timescales (see discussion of energy prices above). However, in
this case the pressure built up over many years and it was not a single event, such as the
financial crisis, that led to the policy changes which took place in 2008.
7.1.2 Friction
As outlined in chapter 2, Jones and Baumgartner (2005) provide two causal mechanisms of
friction: cognitive limits of organisations and institutional procedures.
Cognitive limits of organisations result in disproportional information-processing. Because
organisations can only process a limited number of issues, most issues do not get attention
of major policy makers most of the time. Therefore, the ‘cognitive architecture of
institutions creates an inevitable bottleneck of attention’ (Walgrave and Vliegenthart 2010,
p. 1149).
'Resistance built into the institutional structure of policymaking [my emphasis]’ (Jones
and Baumgartner 2012, p. 4) is a second reason for friction. Past legacies of institutions
can add friction due to sunk costs, long-term budgetary commitments, and bureaucratic
inertia. Therefore, political institutions are comparable to ‘rusted gears’ in the sense that
some force is required to make them move (Workman et al. 2009, p. 80). This makes the
reallocation of resources to other areas more difficult and the response is slow
(Baumgartner et al. 2009).
243
Cross-case analysis
Recent studies also suggest that there may be other sources of friction for example
political parties (Jensen 2011; Walgrave and Varone 2008). This thesis identified an
additional causal mechanism which can explain the phenomenon of friction: the role of key
personnel in organisations which will be elaborated on below.
Friction resulting from changes in key personnel
Changes in key personnel in the SO and the CBRP impacted on the degree of friction. This
observation fits with other studies that stress the role of key personnel for driving or
inhibiting policy change (Andeweg 2000; Chabal 2003; van Asche 2012).
In Germany, tied in with other drivers such as budgetary constraints, the change of
personnel at the ministerial level in late 2009 had an impact on the CBRP when Peter
Ramsauer (CSU) succeeded Wolfgang Tiefensee (SPD) as minister of the BMVBS. Of
course, one would expect some alteration of priorities following the change in government
(Jensen 2011), but Ramsauer’s appointment appeared to have an impact far beyond party
politics. As illustrated in section 5.2.5, Ramsauer opposed the expansion of the CBRP and
even proposed to curtail the CBRP’s funding significantly, even though he faced
opposition from his own party and ministry. The opposition from within his own ministry
and party shows that Ramsauer did not simply represent an official party or ministry
position, but rather followed his own agenda on the matter. Interviewees described him as
a ‘transport minister’ primarily interested in transport infrastructure but not in buildings.
One should, however, also take into account the on-going debates at the time around
budget constraints and the debt brake. Ramsauer linked his arguments to this debate and
embedded the cuts in a wider discussion about public spending. It seems that, while
impossible to prove, a different minister with more balanced priorities giving more
consideration to the overall coalition parties’ positions would probably have reacted
244
Cross-case analysis
differently and have been more open toward the CBRP’s expansion and would have tried
to prevent the cuts to its budget. Therefore, the minister constituted a source of friction in
the sense that he actively blocked the expansion of the programme. The role of a minister
in slowing down or increasing the speed of policy change can be observed in other cases as
well and the academic literature suggests that ‘ministers do make a difference because they
develop their own style at the head of the ministry’ (Chabal 2003, p. 45).
Also in case of the SO in the UK changes in key personnel contributed to friction.
However, while in Germany it was the minister who inhibited further change, in Britain
the director of the gas market regulator blocked an expansion of the SO to the gas sector.
This process can also be understood by applying the concept of friction: the Director
General of OFGAS, Claire Spottiswoode, was supposed to establish domestic sector
energy efficiency programmes in order to help with achieving the national carbon
reduction targets set by government (see section 4.2.5). However, she acted as a bottleneck
blocking most of the pressure put onto the regulator. As a result, policy inputs (such as
demands to promote energy efficiency in order to reduce carbon emissions) were not
translated into policy outputs (in the form of an effective E factor or a SO for the gas
sector, see section 4.2.2). Hence Spottiswoode’s objection was a source of friction leading
to the accumulation of pressure to reduce carbon emissions: EST frequently raised
concerns that without the promised funding from the E factor or a SO in the gas sector it
would be unable to generate the carbon emission reductions of the scale required. The fact
that shortly after his appointment McCarthy implemented the SO in the gas sector
indicates that at the time Spottiswoode was the primary source of friction rather than
cognitive limits or constrained institutional capacities as theorised by PET. However, as
the next section will show, there were also significant institutional constraints which
caused substantial friction.
245
Cross-case analysis
The observations made indicate that the concept of friction could be enhanced by
encompassing the role of key personnel in addition to the other causal processes already
theorised by PET as outlined above. One might, however, argue that the appointment of
new personnel with different priorities might simply represent a shift in organisational
priorities. However, in the case of Ramsauer’s appointment criticism came even from
within his own department and from his own party as illustrated in section 5.2.5. Also
Claire Spottiswoode did not seem to simply mirror the priorities of OFGAS at the time.
Shortly after the new director was appointed, OFGAS changed its approach and became
more receptive for the SO in the gas sector again which it also was before Claire
Spottiswoode’s term as Director General (see section 4.2.2).
Friction resulting from limited institutional capability
Cortell and Peterson (1999) point out that institutional structures may prevent policy
makers from implementing changes in public policies. Hence it depends on the
institutional capability of a system whether policy change takes place or not. Institutional
capability refers to ‘how institutional arrangements create opportunities for, or place limits
on, an official's ability to translate her desire for structural change into policy’ (ibid., p.
190). In other words, the institutional capability determines to what extent policy inputs
can be translated into policy outputs, which is precisely how institutional friction is
theorised in PET (Jones and Baumgartner 2012).
In both cases there is evidence for the existence of institutional friction. In Britain, the
allocation of the target setting powers to the regulator resulted in many years of deadlock
due to a limited institutional capability. In Germany, the financial architecture of the
CBRP and its dependence on public expenditure was the cause of institutional friction.
246
Cross-case analysis
In the 1990s, the SO was not only administered by the regulator, as is the case today, but
also designed by them. Both regulators argued that a substantial increase of the SO’s scale
would be a matter for the Central Government to decide given the implications with regard
to energy bills. The regulators compared the SO to a tax, which would not be something
they felt competent introducing (see section 4.2.3). Due to the inability of the institutions
in place (in this case OFGAS and OFFER) to raise the target, pressure to generate the
carbon emission reductions elsewhere accumulated. Directing the target-setting powers to
Central Government instead of the regulator was seen as the only way of dealing with the
pressure to achieve carbon emission reductions in the housing stock. Analysed through the
lens of friction, the institutional constraints (unwillingness and inability of OFGAS and
OFFER to increase efforts beyond a low level) limited the system’s ability to translate
policy inputs (pressure to cut domestic carbon emissions) into policy outputs (effective
SO). In order to realise substantial carbon emission reductions in the domestic sector
modifying the institutional framework appeared to be the only option and the accumulation
of pressure eventually led to reform. Hence the institutional change that took place was a
deliberate means to achieve carbon reduction goals in the housing sector rather than an
unrelated coincidence.
Historically, the German CBRP had been funded by public expenditure coming out of the
overall annual budget; only recently a partial funding mechanism was introduced
independent of the budget (see section 5.2.4). As any other budgetary item the CBRP
requires parliamentary approval and the responsible government department does not
possess direct decision-making powers. There is a cascade of bureaucratic procedures that
need to be followed when budgets are designed and approved. Historically, budgets had
been built bottom-up. First, the government departments propose a fiscal plan (called
Referentenentwurf) allocating funding to the respective areas of their responsibilities (in
247
Cross-case analysis
case of the CBRP this is the BMVBS). The next step is to reconcile the proposal with the
preferences of other government departments. The Treasury has the responsibility to draft
a proposal taking into account the preferences of the various departments and other fiscal
criteria. Based on the draft by the Treasury, the government agrees on a so-called cabinet
proposal (Kabinettsentwurf) which is then presented to the Bundesrat, a council
representing the 16 Länder (federal states) at the federal level. The Bundesrat discusses the
draft budget and submits proposals for amendments to government. After finalising the
draft budget the government forwards it to the Bundestag, the parliament, which holds a
vote on the budget. The Bundestag refers the draft budget to its various committees for
further amendments – the CBRP’s budget is mainly discussed by the Budget Committee
and the Construction Committee. The Budget Committee has the overall responsibility to
draft the annual budget and needs to consider any requests for amendments. With regard to
the CBRP, the Construction Committee cannot directly make amendments to the funding
proposals. However, the political parties usually submit the same amendment proposals to
both committees at the same time. Once the Budget Committee reaches an agreement, the
draft budget is discussed in the Bundestag and often referred back to the Budget
Committee with additional proposals for modifications. Eventually, the budget is presented
to the Bundestag for a final vote.
Within this rather complicated process the original proposals can change significantly.
Compared to the SO, the legislative procedures which need to be followed when
calibrating the policy instrument are more complex. In Great Britain, DECC can make a
decision about the SO target by setting out secondary legislation (under powers given to
them by primary legislation) that requires parliamentary approval, but parliament can only
outright reject or accept the proposed secondary legislation. Rejection happens very rarely
- the last time a piece of secondary legislation was not approved by the House of
248
Cross-case analysis
Commons was in 1969 (House of Commons Information Office 2008). Hence the
parliamentary approval of secondary legislation on the SO is mainly a formality with no
political significance. In contrast, the CBRP’s budget is set as part of the overall budget
process requiring consent from multiple institutions such as the Treasury, the Cabinet, the
Budget Committee of the Bundestag, and the Bundestag itself. However, no evidence
could be identified that illustrates an impact of these structures on the CBRP’s funding.
Friction as a result of cognitive limits
As outlined above, PET theorises that friction arises from cognitive limits of organisations
(Jones and Baumgartner 2005, 2012). However, this thesis found no evidence in both cases
supporting this hypothesis. This may relate to the difficulties of investigating such a
phenomenon with the methods chosen. Hence cognitive limits could have played a role for
the evolution of the two programmes. It would be a worthwhile question for further
qualitative research to explore the role of cognitive limits for friction.
7.1.3 Policy feedback
‘Because things are the way they are, things will not stay the way they are’ (quoted in
Cook 2007, p. 390)– this quote by the German playwright and theatre director Berthold
Brecht sums up the principle behind policy feedback nicely. The status quo is never stable
and consequences of the status quo, even though they may take years or even decades to
develop, eventually lead to transformation. The conceptual framework distinguished
between two such processes: those that dampen and inhibit expansion of the existing
policy regime and those that drive it.
249
Cross-case analysis
Negative policy feedback
Chapter 2 introduced the concept of negative policy feedback. Following Weaver (2010),
this has been conceptualised as consequences of a policy that undermine, rather than
reinforce, the persistence of this very policy.
There is evidence in both countries for negative policy feedback. However, in Germany
the evidence is more profound. In Germany, the CBRP was subject to fiscal pressures that
had a negative impact on the CBRP’s budget. A good example is the expansion of the
CBRP’s budget in
8 (see section 5.2.3). In order to help with recovery of the struggling
economy after the financial crisis, the German government decided to increase the budget
of the CBRP substantially as part of the first Economic Stimulus Package. However, the
large amount of public spending and the required additional debt could only be justified
with a commitment to limit the budget deficit in the future, which resulted in the
introduction of the debt brake (Dietrich 2009). In 2010, the debt brake had a visible impact
on the CBRP and the responsible minister announced significant cuts of the CBRP’s
budget justified with the debt brake. This is an interesting case whereby an expansion of a
policy instrument driven by positive policy feedback processes had knock-on effects in the
form of negative policy feedback processes. In order to provide more long-term stability
with regard to the CBRP’s financial resources a budget-independent Energy and Climate
Fund was introduced subsequently. However, it looks like the introduction of the fund may
have caused the opposite as it was based on rather unreliable financial sources itself.
Again, even though unintended, the modified finance structure started to generate negative
policy feedback effects in the form of unreliable funding for the programme.
As indicated above, the SO was not subject to negative policy feedback to the extent the
CBRP was. There were at least two dynamics that can be classified as negative policy
250
Cross-case analysis
feedback. Using energy suppliers as a vehicle to promote energy efficiency measures
(probably inevitably) resulted in resistance from the obligated parties. Throughout the
history of the SO attempts have been made by suppliers to weaken the targets or change
the rules in their favour (see section 4.2.8). Resistance to existing policies from negatively
affected vested interests constitutes one form of negative policy feedback (Schrad 2010).
However, even though some observers predicted significant obstacles in the form of
resistance from the incumbent energy industry against a utility energy efficiency
programme (Dawkins 1987), this kind of policy feedback proved fairly weak and positive
policy feedback in the shape of the targets consistently being achieved more than
outweighed the negative policy feedback. Eventually, Government did pay less and less
attention to the energy suppliers’ concerns based on the experience that they always
delivered the targets (they did not at the end of CERT but this is outside the scope of the
thesis because it happened after 2010).
Another dynamic representing negative policy feedback is the circumstance that most of
the cost of the SO is passed through to households via the energy bills. This led to
concerns around social equity and its impact on the fuel poor with some deeming the SO
as a policy with regressive impacts (Boardman 2010), which it clearly is one the revenue
raising side. In order to address such concerns, the SO contained provisions for directing a
significant proportion of the effort towards households in fuel poverty, although this also
was very controversial (see section 4.2.6). However, even though the SO had regressive
effects due to the way revenues are raised, there was no evidence showing that it had a
negative impact on the overall size and ambition of the policy instrument i.e. that
Government deliberately did not expand or scale down the SO in response to those
concerns.
251
Cross-case analysis
Positive policy feedback
Based on the historical institutionalism literature (Campbell 2002a; Mettler and Soss 2004;
Pierson 1993b, 2004a; Skocpol 1992) and PET (Baumgartner and Jones 1991, 1993,
2002a, 2002b, 2009; True et al. 2007a), chapter 2 introduced the concept of positive policy
feedback defined as mechanisms that stabilise and reinforce existing policy regimes.
Such processes are apparent in both cases, the SO and the CBRP. The SO provided
significant funding for energy efficiency measures and the energy efficiency industry
undoubtedly benefited substantially from the SO. Markets such as the cavity wall
insulation sector grew rapidly over the last decade as a result of the incentives put in place
by the policy instrument. With growing market size, the number of jobs in home energy
efficiency retrofits increased and thereby created interest groups that are heavily dependent
on the SO. As discussed in chapter 2, such effects whereby a policy creates new interest
groups which then support the maintenance and expansion of this policy are well-known in
the literature on policy feedback (Campbell 2002a; Campbell 2002b; Pierson 1993a;
Skocpol 1992).
Policies do not only provide resources for particular interest groups, they can also help the
organisation and formation of interest groups (Pierson 1993a) because ‘[p]rogram[me]
resources also create material incentives for groups to band together in political activity’
(Mettler and Soss 2004, p. 62). In the UK energy efficiency industry lobbying
organisations can trace their roots back to the 1980s and expanded subsequently as a result
of a growing industry. Organisations such as ACE, NIA, and BEEF managed to get
involved more directly in the decision making processes, for example by being consulted
prior to the official consultation stage and being part of important stakeholder forums such
as EEPH.
252
Cross-case analysis
While maybe not as visible as in the UK, the CBRP also created powerful interest groups
and
associations
representing
the
construction
industry,
housing
corporations,
homeowners, and tenants. Expansion of the CBRP was generally supported by these
groups and budget cuts opposed. Particularly the announced reductions of funding for the
Programme in more recent years triggered a response from interest group organisations
trying to make the case for stopping the contraction of the CBRP. However, historically
the German energy efficiency industry was not as well organised as their British
counterpart. Only in 2010 was the German Industry Initiative for Energy Efficiency
(Deutsche Unternehmensinitiative für Energieeffizienz – DENEFF) founded; the leading
UK organisation, the Association for the Conservation of Energy, can trace back its roots
to 1981 (Pickvance 2009). In the UK, the expansion of the energy efficiency industry as a
result of the SO and with it an increasing lobbying power certainly did contribute to
maintaining and enlarging the SO. The picture in Germany is quite different: since the
mid-1990s the construction industry, and as part of it the energy efficiency industry
focusing on building refurbishment, had been in decline (see section 5.2.2). One would
expect a negative effect on the CBRP assuming decreasing lobbying power over time.
However, the struggling construction industry was used as an argument for expanding the
CBRP rather than scaling it down due to the perceived positive impact of the Programme
on employment and economic activity, which constitutes another form of positive policy
feedback.
The different responses to labour market dynamics can be explained by Esping-Andersen’s
(1990) typology of welfare regimes classifying states in three distinct groups - liberal,
conservative and social democratic regimes.
ermany and the UK are considered ‘ideal’
types falling into the conservative and liberal category respectively. According to this
classification, Germany puts emphasis on maintaining its status quo whereas the UK
253
Cross-case analysis
employs a laissez-faire approach leaving it to the market forces. While probably too
stylised, this tendency can be observed when looking at past active labour market policy:
using public policy for job creation has a long legacy in Germany (Sommer and Rosenthal
2012). Historically, Germany allocated significantly more public expenditure to active
labour market programmes than the UK (Bonoli 2010; Kluve 2010). According to Chung
and Thewissen (2011), in times of economic slowdown Germany tends to employ labour
market policies that are intended to keep the workforce in the labour market in order to
preserve their skills, an approach that is seen as crucial in Germany to maintain its
comparative advantage (Hall and Soskice 2001). The UK appears to rely more on market
forces and displays a low degree of active government intervention with regard to labour
market policy. This became particularly apparent after the financial crisis in 2008 when
Germany introduced two economic stimulus packages (which included an increase in the
CBRP’s funding, see section 5.2.3). The UK, in contrast, did not implement any measures
comparable to the German response (Chung and Thewissen 2011) relying on the
continuation of the past growth regime (Clegg 2010). This is also mirrored in the history of
the SO: arguments around the positive effects of the SO on the labour market or industry
did not play an important role if any for the development of the policy instrument.
To conclude, in Great Britain and more muted in Germany, the SO and the CBRP did
create new interest groups associated with increased lobbying power, which in turn put
pressure on the Government to maintain and expand the policy instrument. It is a classic
example of positive policy feedback according to the theoretical literature (Pierson 1993a).
Additionally, the perceived positive effects of the CBRP on employment and economic
activity, and the recognition that policy makers should make use of such instruments to
stimulate the economy and the labour market fall into the category of ideational and
symbolic legacies, another form of positive policy feedback (Béland 2010).
254
Cross-case analysis
7.1.4 Summary
In both cases, many of the processes of policy change can be understood through a lens of
external pressures, friction, and policy feedback. The following table categorises the
observed processes according to the previous sections.
Table 12: Summary of policy change mechanisms identified
Mechanism
External pressures –
subsystem spillovers
Germany – CBRP
Climate change policy
External pressures – systemic
perturbations
Friction – key personnel
Financial crisis
Friction – institutional
structures
Friction – cognitive limits
Positive policy feedback
Negative policy feedback
Great Britain – SO
Climate change policy
Rising energy prices
Fuel poverty
Resistance from Construction
Minister
Budget process
Resistance from Director General of
the Regulator
Target setting powers with Regulator
Perceived positive employment
effects
Negative fiscal effects
Growing energy efficiency industry
Perceived successfulness
Resistance from energy suppliers
Regressiveness of SO
Interestingly, there are idiosyncratic as well as common drivers. For example, fuel poverty
has been an important driver in the UK, but, until recently, has barely been mentioned in
Germany. The reason for this is not that fuel poverty does not exist in Germany, it does
and in more recent years more attention is paid to this issue, even though data availability
is poor (Kopatz et al. 2010). The welfare system in Germany does not accommodate for
rising energy bills either. As things stand, there is no convincing explanation for this
remarkable difference.
Also, there are unexpected ‘non-drivers’ such as the influence of EU policy. Neither of the
two case studies indicated that EU policy had an impact on the SO and the CBRP. The
reason for this may be that both Germany (Lees 2007a) and the UK (Jordan 2004) are now
considered EU members that ‘upload’ their own political agenda to the European level
rather than following EU directives. However, Germany has been reluctant to adopt
255
Cross-case analysis
market-based instruments (Lees 2007a), such as the SO, and transposition of existing
building directives into national law in the UK has been somewhat muted (Ekins and Lees
2008). Potentially, European policy could become more important in the future with the
adoption of the new EU Energy Efficiency Directive.
The next step is to evaluate to what extent the hypotheses developed in chapter 2 are
supported by the evidence reviewed so far.
7.2 Hypotheses
In chapter 2 four hypotheses were presented in order to focus the research on a number of
key issues.
7.2.1 Hypothesis 1
The first hypothesis related to the patterns of policy change resulting from long-term
pressures:
H1/1: Long-term gradual pressures lead to changes of policy instruments that
follow a pattern of punctuated equilibrium. The scale of change is substantial
compared to the status quo.
PET predicts that policies remain relatively stable for long periods of time followed by
short periods of rapid, dramatic change.
The rival hypothesis to this is that policies change gradually in line with the pressures,
keeping a balance between the policy inputs and outputs:
H1/2: Policy instruments change incrementally as a result of gradual pressures.
The scale of change is minor compared to the status quo.
With regard to hypothesis H1/1, the following observations can be made: neither of the
two cases showed policy change along the patterns of punctuated equilibrium.
256
Cross-case analysis
Methodologically, PET scholars plot policy changes as a histogram and overlay it with a
Normal curve with similar standard deviation. However, the number of changes measured
in this thesis is <20 and not suitable for such an approach. Rather, a qualitative assessment
of the degree of punctuation is required (Baumgartner, personal comm. 05/10/2012).
In case of the SO, the energy savings targets can be used as an indicator of the scale of
policy change. Using the settings of policy instruments as an indicator of policy change is
common practice (Hall 1993; Howlett 2007; Howlett and Cashore 2009). A good metaphor
to describe the pattern of past SO targets is a staircase with steps getting higher and higher,
but not that of punctuations and equilibrium. For the CBRP, the budget of the programme
can be used as an indicator of policy change; budgets are often used as proxies when
investigating policy change in PET (Breunig and Koski 2006; True 2000). Again, the
development of the CBRP’s budget over time follows a more gradual pattern of smaller
changes or steps, but punctuations are not apparent.
However, H1/1 proves correct in the sense that ‘the scale of change is substantial
compared to the status quo’. In both cases there has been significant change. The SO
targets increased by a factor of almost 70 from 1994-1998 to 2008-2012 (annual lifetime
savings) (see section 4.1.2). While less profoundly, the CBRP’s budget increased tenfold
between
1 and
9. There was a dip in
1 , but the budget going forward of €1.5
billion is an eightfold increase compared to 2001 (see section 5.1.2). The scale of change,
particularly in case of the SO, is certainly substantial compared to the status quo.
Hypothesis H1/2 predicted incremental and minor change. In the cases observed in this
dissertation, the magnitude of change observed exceeds the ranges usually considered as
incremental. There is no widely acknowledged cut-off point beyond which change is
considered non-incremental, making it difficult to refute the claim that all change is
257
Cross-case analysis
incremental. With regard to budgets, a threshold of 10% has been suggested as the cut-off
point (LeLoup and Moreland 1978) and is used by Anderson and Harbridge (2010) in a
recent study reviewing the range of cut-off points in the literature.
Using the 10% cut-off point, the changes observed in case of the CBRP fall into the nonincremental category with an increase close to 20% on average per year. The SO also
exceeds the 10% threshold showing almost a 30% increase on average per year.
In summary, hypothesis H1/1 is only partially supported by the evidence in the two cases
analysed in that change is substantial compared to the status quo, but not in that the pattern
of change are punctuated. Hypothesis H1/2 is not supported by the evidence because the
scale of the observed change is not incremental as predicted.
The patterns of gradual but profound policy change is in line with observations by others
that policy outcomes are not punctuated (Cashore and Howlett 2007; Givel 2008, 2012;
Perl and Dunn 2007). Much of the literature using PET focused on the policy inputs for
example in the form of media attention (Givel 2012).
The latest research by Givel studying the 2004 anti-tobacco law of the country of Bhutan
suggests that policy change can indeed be ‘sweeping but nonpunctuated’ (Givel 2012, p.
645). Givel stresses that this is a result of the complex policy process and the interplay of
multiple factors such as the state, corporate actors, interest groups, and the market.
It is noteworthy to mention that the concept of friction developed by PET predicts that the
degree of punctuation depends on the stage in the policy process. Policy inputs such as
media attention are assumed to be highly punctuated whereas policy outputs will be less
punctuated. Hence the further an issue is processed by the policy system (from agenda
setting to implementation), the lower the degree of punctuation. This is driven by cognitive
limits and institutional structures (see sections 2.3.1 and 7.1.2). There is evidence for this
258
Cross-case analysis
phenomenon in recent analysis of PET scholars (Baumgartner et al. 2009; Jones et al.
2009; Jones et al. 2003; Walgrave and Vliegenthart 2010).
To conclude, the results of the two case studies that policy outputs can be substantial, but
not necessarily punctuated, are supported by a body of literature critical of PET as well as
the notion of friction established by PET itself.
7.2.2 Hypothesis 2
The second hypothesis is related to friction:
H2/1: Friction significantly limits the capability of institutions to respond to
gradual pressures and creates a misbalance between policy inputs and policy
outputs. The result is the accumulation of pressures.
The rival hypothesis is that there may as well be no evidence for friction of any kind and
policy inputs are linearly translated into policy outputs.
H2/2: Friction does not limit the capability of institutions to respond to gradual
pressures and there is a balance between policy inputs and policy outputs.
Pressures do not accumulate as a result.
Both cases displayed the occurrence of friction at various points in time (see section 7.1.2).
Therefore, hypothesis H2/1 is supported by the evidence whereas H2/2 is not. Instances
where friction can be observed are discussed in the following section.
7.2.3 Hypothesis 3
Given that in both case studies there is evidence for friction, the next hypothesis
conceptualizes the two causal mechanisms that lead to friction according to PET:
259
Cross-case analysis
H3/1: Friction is caused by a) the limited ability of institutions to deal with
information i.e. policy inputs, and b) prohibitive institutional structures that do not
allow for effective translation of policy inputs into policy outputs.
Theoretically it is possible that other factors than those proposed in hypothesis 3 cause
friction resulting in alternative hypothesis 3/2:
H3/2: Friction is caused by other factors.
Section 7.1.2 illustrated that there is ample evidence for friction caused by prohibitive
institutional structures. In case of the SO, the responsibilities for target setting lay with the
regulator for a long time essentially causing an impasse in terms of policy change. The
regulator refused to expand the targets arguing this was outside its remit. Only when the
Government took over the target-setting process did significant policy change occur, a
clear indication that the institutional structure did not allow for a substantial modification
of the SO (see section 4.2.3).
The budget of the CBRP is subject to parliamentary approval and the responsible
government department does not possess direct decision-making powers. The cascade of
bureaucratic procedures that needs to be followed makes rapid and dramatic modification
difficult to facilitate, even if the department in charge of designing the CBRP supports
such change (see section 7.1.2). Procedures like this add to the degree of friction within the
policy process.
As discussed before, establishing whether cognitive limits cause friction is difficult to
establish - in both cases the evidence was not sufficient to either validate or refute the role
of the constraint ability of institutions to process information. However, this does not mean
that cognitive limits did not play an important role. Further research investigating the role
260
Cross-case analysis
of cognitive friction at the level of organisations could systematically assess to what extent
cognitive limits have an effect on the degree of friction.
Therefore, hypothesis H3/1 is supported in that it attributes friction to prohibitive
institutional structures, but the predicted impact of cognitive limits cannot be validated or
falsified due to a lack of data.
Hypothesis H3/2 proposed that other factors than the two suggested by hypothesis H3/1
may cause friction. H3/2 is correct in the sense that for both cases an additional causal
mechanism leading to friction has been identified, but false in that it disqualifies the
impact of prohibitive institutional structures. As section 7.1.2 demonstrates, in both cases
changing key personnel played a pivotal role in the history of policy change and caused
friction. Probably more profound than in case of the CBRP, the role of Claire
Spottiswoode as Director General of OFGAS demonstrates how changing key personnel
can increase or decrease the degree of friction. Similarly, the political priorities of Peter
Ramsauer, the Minister of the BMVBS in charge of overseeing the CBRP, slowed down
the CBRP’s expansion in terms of its budget and almost resulted in budget cuts of 50% in
2010.
Integrating the role of key personnel in the friction concept would, at least based on the
two case studies analysed here, improve its explanatory power. The academic literature
also indicates that focusing on the role of key personnel can enrich analysis of public
policy (Andeweg 2000; Chabal 2003; van Asche 2012).
7.2.4 Hypothesis 4
Hypothesis 4 relates to feedback mechanisms, including positive and negative policy
feedback. H4/1 proposed that:
261
Cross-case analysis
H4/1: Existing policy instruments create positive and negative feedback processes.
Positive policy feedback leads to an expansion of the policy whereas negative
policy feedback results in a dampening of the expansion or contraction of the
policy.
Alternatively, H4/2 suggested that politics independent of the policy instrument itself
determine its modification:
H4/2: Existing policy instruments do not create positive and/or negative feedback
processes. Policy change is a result of political processes not influenced by
existing policies.
The analysis above clearly supports hypothesis H4/1 and refutes hypothesis H4/2. Policy
feedback occurred in both cases. Positive policy feedback includes the perceived positive
employment effects (CBRP), the growing energy efficiency industry (SO and to some
extent the CBRP), and the perceived success of the policy instrument (SO). There is also
evidence for negative policy feedback in the form of negative fiscal effects (CBRP),
resistance from energy suppliers (SO), and the regressive impact of the policy instrument
on energy bills (SO).
7.3 Limitations of the analysis
The analysis carried out potentially suffers from at least two weaknesses. One is
methodological, the other conceptual. The methodological weakness stems from the fact
that case studies employing systematic process analysis based on interviews potentially
suffer from various biases that may skew the results (Tansey 2007). As outlined in chapter
3, document analysis and academic literature were used as a means of triangulation.
However, in some instances the analysis relies mostly on interviewees’ responses
262
Cross-case analysis
particularly in areas where the interpretation of the interviewee could not be cross-checked
with other forms of evidence.
A conceptual limitation is the treatment of the various causal mechanisms and policy
drivers which cause policy instrument change. For the purpose of the analysis, the different
mechanisms and drivers have been discussed separately. It is, however, very difficult to
determine the relative importance of the causal mechanisms and the various policy drivers
due to difficulties of measuring their relative importance. Also, the separate treatment of
the mechanisms and drivers, which was chosen for analytical purposes, is somewhat
artificial and obscures the fact that they are interrelated. One driver could, for example,
reinforce (for example climate change policy and the positive effect on the energy
efficiency industry) or undermine (for instance in case of the CBRP negative fiscal effects
resulting from increased spending) another driver. Where possible, the connectedness has
been discussed, but there may have been instances where the analysis missed links
between the drivers and mechanisms.
However, one can be confident in the broad results of the analysis. With regard to the
causal mechanisms, in both cases there was substantial evidence that external pressures,
friction, and policy feedback played a major role for the policy changes observed. Also, an
additional mechanism of friction showed in both the UK and Germany indicating that it
may operate elsewhere too. Further research, as outlined in the next chapter, could help
exploring the role of friction further and increase the confidence in the proposed
mechanism.
263
Conclusions
8 Conclusions
If a man will begin with certainties, he shall end in doubts: but if he will be content to begin with
doubts, he shall end in certainties.
Sir Francis Bacon (quoted in Wormald 1993, p. 39)
The purpose of this last chapter is to step back from the detail presented in the previous
chapters and to locate this thesis in the wider energy literature and the theories of policy
change.
I do this by answering the initial research questions posed in chapter 1 and summarising
this thesis’ contribution to the academic literature. Furthermore, this chapter suggests
avenues for further research and discusses policy implications.
8.1 Answering the research questions
Each of the research questions stated in the introduction is reconsidered below.
1) How did the key domestic energy efficiency policy instruments in the UK and
Germany change since their inception?
Overall, both the SO and the CBRP changed significantly in terms of scale, scope, and
structure. In terms of scale this can be seen by looking at the target size of the SO and the
funding base of the CBRP. The SO’s targets increased by a factor of almost 7 from 199 1998 to 2008- 1 (annual lifetime savings) and the CBRP’s funding was raised from
about €1
million to €1,5
annually post-2010 (see sections 4.1.2 and 5.1.2). In both
cases the scope in terms of the eligible measures evolved as well. The SO increasingly
focused on insulation measures and allowed new technologies and even behavioural
264
Conclusions
measures into the scheme. The CBRP moved from strictly defined packages of measures
towards setting an energy performance standard based on the Energy Saving Ordinance.
The structure of both programmes changed too. Since 2002, the target for the SO is set by
the responsible Secretary of State rather than the regulator. Historically, the CBRP’s
funding was based on the budget, now an additional, budget-independent mechanism
provides a large share of its resources.
2) What kind of pressures were drivers of policy change?
There were many pressures that acted as drivers for policy change in the case of the SO
and the CBRP. I focused on those pressures seen as key drivers by interviewees and
prevalent in the documents analysed. Both cases showed that climate policy and the
perceived impact of climate change was a key pressure particularly in the last decade. In
case of the SO, in the early days, climate policy played a lesser role. Instead, energy
market liberalisation, and with it the idea of Least-Cost Planning, triggered the
implementation of a policy that used utilities as a vehicle for delivering energy efficiency
measures. The
erman CBRP’s inception was clearly driven by climate policy
considerations, but this was aligned with labour market policy objectives given the decline
of the German construction industry since the mid-1990s and because the CBRP was seen
as an employment-generating instrument. Labour market dynamics and the economic
climate in general also came to the fore at other times, for example during the financial
crisis and the following recession, when the CBRP was explicitly used as part of an
economic stimulus package. The SO was also driven by rising energy prices and
increasing fuel poverty, the former resulting in a significant uplift of the energy savings
target of the SO in 2008. Although the SO was never intended or explicitly used to reduce
265
Conclusions
fuel poverty, the social equity provisions and their maintenance over time are clearly
linked to pressure on the government to alleviate fuel poverty.
3) How did the principal policy instrument of the policy subsystem of domestic
energy efficiency policy respond to gradual pressures?
Even though the scale of change was remarkable, both policy instruments changed
gradually over time and, on the whole, not abruptly. Both policy instruments showed
periods of stagnation as well as times of faster change. Times of stagnation in the case of
the SO can be found particularly in the first 8 years of the policy instrument when the
regulator resisted an expansion of the scheme. More recently, the CBRP was subject to
budget constraints and its expansion slowed down with no foreseeable increase in the
scheme’s financial resources. More rapid change happened in the case of the SO after the
Secretary of State had been allocated the responsibility for setting the target in the 2000
Utilities Act. Large increases in the target could be observed from EESoP 1 to EEC 1
(2002), EEC 1 to EEC 2 (2005), and EEC 2 to CERT (2008) - more or less doubling the
target every time. Furthermore, during 2008, on the back of the discussions around a
potential windfall tax on energy suppliers, the CERT target was increased by 20%.
Substantial increases in the funding of the CBRP took place in 2003 when the budget was
almost doubled, in 2005 when the new coalition government decided to allocate about
three times more funding per year, and in 2008 when the budget was increased again as
part of the economic stimulus measures. Overall, there was a pattern of increasingly large
changes with each shift in the program (although the CBRP’s expansion slowed down after
2009).
4) How well can the concept of friction explain these processes? If there is evidence
for friction, which factors caused it?
266
Conclusions
The evidence provided in chapters 4, 5, and 7 shows that friction can indeed explain many
of the changes observed. PET conceptualises two sources of friction - prohibitive
institutional structures and cognitive limits. This thesis illustrated the effect of prohibitive
institutional structures on policy change, mainly in the case of the SO prior to the transfer
of target-setting powers to the Secretary of State, but also in the case of the CBRP which
has been dependent on bureaucratic budget setting processes involving multiple
institutional hurdles. Friction resulting from cognitive limits and the inability of
organisations such as government departments and agencies to process incoming
information proportionately, has proven difficult to investigate within this research project.
To what extent cognitive limits played a role therefore remains an open question. Both
case studies suggested that key personnel can, under some circumstances, constitute a
source of friction that goes beyond mere institutional structures. In the case of the SO,
Claire Spottiswoode, former Director General of OFGAS, deflected political pressure to
expand the SO primarily driven by personal preferences as illustrated in sections 4.2.2 and
7.1.2. Similar patterns can be observed for the CBRP - Peter Ramsauer, Minister for
Transport, Building and Urban Development, attempted to scale down the Programme due
to his personal priorities even though his own government department and party were in
favour of maintaining its funding base. So far, the literature on friction does not account
for such mechanisms.
5) Which role did policy feedback processes play for policy change?
There is ample evidence for policy feedback processes in both cases. I distinguished
between two types of policy feedback processes – negative and positive policy feedback.
Negative policy feedback processes included negative fiscal effects of increased funding
for the CBRP; increased funding contributed to public expenditure at a time when public
debt was becoming a political issue, revenue raising under the SO had a regressive
267
Conclusions
character, there were increasing costs that had to be paid by consumers, and, to a limited
extent, there was resistance from energy suppliers to increasing the SO targets. Positive
policy feedback was evident in the form of a growing energy efficiency lobby benefiting
from the SO and the CBRP, the perceived positive labour market and economic stimulus
effects of the CBRP, and the perceived success of the SO. Overall, positive policy
feedback probably outweighed negative policy feedback given the significant expansion of
the programmes.
8.2 Contribution
This thesis started from the presumption that politics are underplayed in the current body
of literature on energy efficiency policy. I argued that most scholars focus on technological
questions, the cost of different technologies, and the economics of policy instruments, all
of which are important research areas. However, the reality of how policy instruments are
implemented and modified over time does not simply follow such factual information. If it
is not just ‘facts’ that cause policy change but political factors, a detailed analysis of the
politics driving energy efficiency policy can illuminate those mechanisms.
Understanding under which conditions policy change is likely to be feasible can help a
great deal when it comes to future policy design. Too often policy proposals are doomed to
failure because the constraints and drivers of policy change are not paid sufficient
attention. Furthermore, using a comparative approach and looking at two countries that
both have implemented ambitious policies to reduce domestic all energy consumption
offers potential for policy learning, lesson-drawing and policy transfer.
More specifically, this thesis made the following contributions:
268
Conclusions
first, this thesis not only demonstrated that politics affected the two policy instruments
over time, but also which mechanisms more generally were responsible for the changes
observed.
Second, it provided the first comprehensive account of the history of both the CBRP and
the SO, two policy instruments which attract a lot of attention both nationally and
internationally. Such an historic perspective enables other scholars to understand where
those policies were coming from, how they changed, and why they ended up the way they
did.
Third, this thesis employed a relatively novel theoretical framework based on the concept
of friction coming out of PET and the more recent policy feedback literature. The analysis
in chapter 7 identified areas for potential extension of those theories in future revisions.
Finally, by contrasting the two cases, I demonstrated similarities and differences between
German and British energy efficiency policy.
8.3 Policy implications
This thesis started with an observation – that the impact of politics on policy is rarely
considered in the academic literature on energy efficiency policy even though it has long
been recognized in political science that ‘policy determines politics’ (Lowi 1972, p. 299).
However, particularly the underlying funding mechanism plays a key role when it comes
to politics.
Taxpayer-funded energy efficiency policies
Energy efficiency policies based on public budgets are always at risk from future funding
cuts due to fiscal politics. This is partly the result of negative policy feedback in the form
of contributing to increasing public debt but also due to the exposure to external events
269
Conclusions
such as financial crises. In Germany, the CBRP is taxpayer-funded and the Finance
Ministry and the parliament are directly involved in the budget-setting process. During
times of austerity, the CBRP almost always became subject to budget cuts. Ideally, in
order to protect energy efficiency policies that require a large amount of funding the
finance for the policy or parts of it should be provided independently from the overall
budget. However, this is by no means an easy task.
Consumer-funded energy efficiency policies
One could argue that consumer-funded programmes, such as the SO, are better protected
against potential austerity measures. In fact, this was the part of the rationale when the SO
was implemented in Britain: the costs are borne by energy bill payers and therefore the
Treasury has no involvement in the financial transactions taking place. Hence, even
substantial spending cuts as seen at the moment did not affect the SO whereas taxpayerfunded energy efficiency programmes such as Warm Front were terminated. The example
of the SO illustrates that consumer-funded energy efficiency policies may indeed be better
protected from fiscal politics and external pressures.
However, consumer-funded energy efficiency policies create their own politics due to
negative policy feedback: the downside of consumer-funded programmes is that their
revenue raising part is always regressive in a system where costs are passed on to
consumers without considering their financial status. If social equity provisions are not
sufficient, this effect can potentially undermine the programme’s financial stability.
Counterbalancing regressiveness
Those negative policy feedback effect can be counterbalanced either by keeping the scale
of funding required for energy efficiency policies based on energy bills small so the
270
Conclusions
regressive effects are minimal or by providing a large number of measures to the majority
of households and particularly those on low incomes and in inefficient properties. This has
been the approach of the SO in the UK in the early years of the obligation where the
contribution from households was less than £10 per year and most households received an
energy efficiency measure. (at least CFLs).
However, over the years the scale of the SO increased and contributions are now more
than £50 per year per household (see analysis in chapter 4). In addition, more recently, the
UK government decided to reform the SO and in the future energy suppliers will be forced
to deliver mainly high-cost measures including solid wall insulation and hard-to-treat
cavity wall insulation. Such a move has serious equity implications. If the SO is reoriented
towards high-cost measures this results in a smaller number of households who benefit
from the policy instrument and increases the number of those who pay but do not receive
any energy efficiency measures (Rosenow et al. 2012).
Lessons learned for the fiscal implications of consumer-funded energy efficiency
policies
Germany is increasingly looking at energy efficiency policies that can be financed
independent of the public budget. Energy savings obligations could potentially play an
important role for the delivery of low-cost measures. Policy makers considering energy
savings obligations need to consider the limits of such programmes due to negative policy
feedback in the form of regressive effects. Counterbalancing regressive effects by
progressive delivery of measures (i.e. by allocating a large share for households on low
incomes living in low-energy efficiency homes) becomes more difficult when energy
savings obligations are focused on high-cost measures. This is because a smaller number
271
Conclusions
of measures can be financed with the same amount of money leaving more households that
contribute but do not benefit from the policy.
Questions that should be asked when designing energy savings obligations include: what is
the scale of funding required to provide the energy efficiency improvements aimed for in
the long term? What are the implications for energy bills and the impact on fuel poverty?
Which measures can be promoted that benefit the majority of households in order to offset
the regressive impacts? Dealing with those questions early on can avoid some of the
negative feedback consumer-funded policies tend to develop but also illustrate that
additional policies are needed.
Lessons learned for the fiscal implications of taxpayer-funded energy efficiency
policies
Taxpayer-funded energy efficiency policies avoid regressive effects assuming that taxes
are progressive overall. However, they develop negative policy feedback too in form of
increasing public debt and are exposed to the interference of fiscal politics. In theory, one
could use hypothecated taxes to fund energy efficiency policies. This would essentially
ring-fence the finance for those policies and protects them from budget cuts. In Germany,
the ecological tax is partly used to fund the pension system – the CBRP could potentially
also be financed through such a mechanism. Using the revenues from the auctioning of EU
ETS permits proved to be an unreliable source of funding in Germany and did not provide
more certainty with regard to future funding. Similar ideas are currently promoted by the
Energy Bills Revolution in the UK and some important lessons could be learnt from the
CBRP.
272
Conclusions
8.4 Avenues of further research
There are at least two areas where avenues of further research can be found – within the
literature on policy change theory and the cases analysed in this thesis.
Regarding the latter, the timeframe of the two cases covered by this thesis ended in 2010.
However, in the last two years interesting policy changes took place that represent a fairly
radical diversion from the past in many ways: in the SO’s next period the Energy
Company Obligation (ECO) succeeds CERT. Although similar in principle, ECO differs in
many ways from previous obligations. It has a much stronger fuel poverty focus in that it
has a deliberate fuel poverty objective and an affordable warmth sub-target. Most support
will go toward high-cost measures such as solid wall and hard-to-treat cavity wall
insulation, resulting in much lower carbon and energy savings than CERT/CESP. Also,
ECO is supposed to support the Green Deal, a new on-bill finance scheme for building
retrofits, and the two are closely linked (Rosenow and Eyre 2012). Analysing the
implications of those changes and the politics that led to them would be a fruitful area for
future research.
For both policy instruments the issue of funding deserves more research. The CBRP is
now financed through a new funding mechanism, the Energy and Climate Fund. Although
this new funding mechanism was supposed to provide long-term stability in the CBRP’s
budget, it seems to have added, rather than alleviated, uncertainty (Rosenow in press). This
situation is likely to trigger further policy change and the implications of the modified
funding mechanism are worth investigating. Also the SO’s funding mechanism is likely to
be subject to modification. Since 2011/2012, some consumer-funded policies are included
in the Levy Control Framework, essentially a cap on off-balance sheet spending imposed
by the Treasury on DECC (HM Treasury 2011). So far, the SO is not covered by the Levy
Control Framework (DECC 2011b), but there is no reason why it could not be included in
273
Conclusions
the future, particularly in response to concerns about the increasing costs of the SO (NERA
2012). Exploring the limits of utility-funded energy efficiency policies in terms of the
financial burden put on energy bills and the political tensions this involves is an interesting
avenue for more scholarly work.
Unanswered questions also include the varying importance of issues such as fuel poverty
in Germany and Britain – so far no convincing explanation exists for why this area is
treated so differently in the two countries (Kopatz et al. 2010). However, more recently,
there are signs that the debate is starting to pick up in Germany as well. It is also unclear to
what extent EU policies will alter domestic energy efficiency policy in the future given its
limited impact in the past on the SO and the CBRP. The new Energy Efficiency Directive
could have significant repercussions for national policy, for example, the directive
prescribes the introduction of energy savings obligations like the SO in all member states
and Germany considers whether it should do so or make use of the exemption provisions
(Bürger et al. 2012).
Concerning the theories of policy change, this thesis showed that there are a number of
promising questions that could be asked. First, under which circumstances do key
personnel contribute to friction and when do they not? To what extent does this impact on
the institutional structures causing friction and vice versa? Second, how can cognitive
friction be analysed in qualitative research? How significant is its impact? Finally, what
role does the societal and political ‘memory’ of external pressures play? Does it matter
how long the pressure lasts? Do pressures always accumulate if the response is
disproportionate?
274
Bibliography
9 Bibliography
ACE 1985. Lessons from America. No. 2. US gas and electricity utilities and the
promotion of conservation. London: ACE.
ACE. 2002. Energy Policy: Key Issues for Consultation. Response from the Association
for the Conservation of Energy. September 2002 [Online]. ACE. Available:
http://www.ukace.org/publications/ACE%20Response%20%282002-09%29%20%20DTI,%20Defra%20and%20DTLR%20Consultation%20on%20energy%20whit
e%20paper.pdf [Accessed 11/01 2012].
ACE. 2006. Initial consultation by DEFRA: “The Energy Efficiency Commitment – April
2008 to March 2011”. Response by the Association for the Conservation of Energy
[Online].
Available:
http://www.ukace.org/publications/ACE%20Response%20%282006-09%29%20%20Defra%20initial%20Consultation%20on%20EEC3.pdf
[Accessed
11/01
2012].
ACE. 2009. Department of Energy and Climate Change consultation: ‘Amendments to the
Carbon Emissions Reduction Target’. ACE response – April 2009 [Online].
Available:
http://www.ukace.org/publications/ACE%20Response%20%28200904%29%20-%20ACE%20%20%20DECC%20Amendments%20to%20the%20Carbon%20Emissions%20Red
uction%20Target.pdf [Accessed 11/01 2012].
ACE. 2010. DECC Consultation on Extending the Carbon Emissions Reduction Target.
ACE
response
–
March
2010
[Online].
ACE.
Available:
http://www.ukace.org/publications/Consultation%20response%20%28201003%29%20%20DECC%20Extending%20the%20Carbon%20Emissions%20Reduction%20Tar
get.pdf [Accessed 11/01 2012].
Adams, H. 1918. The education of Henry Adams : an autobiography, Boston, Houghton
Mifflin.
AEA. 2009. Carbon factor for wood fuels for the Supplier Obligation [Online]. Available:
http://www.decc.gov.uk/assets/decc/11/about-us/Science/3153-final-report-carbonfactor.pdf [Accessed 15/07/2012 2012].
AEAT 2004. Measurement of the Performance of Cavity Wall Insulation Installed in
Domestic Dwellings. Harwell: AEAT.
A Energiebilanzen 1 . Auswertungstabellen zur Energiebilanz f r die Bundesrepublik
Deutschland 1990. Berlin: AG Energiebilanzen.
Anderson, A. 1991. Source strategies and the communication of environmental affairs.
Media, Culture & Society, 13, 459-476.
Anderson, S. & Harbridge, L. 2010. Incrementalism in Appropriations: Small
Aggregation, Big Changes. Public Administration Review, 70, 464-474.
Andeweg, R. B. 2000. Ministers as double agents? The delegation process between cabinet
and ministers. European Journal of Political Research, 37, 377-395.
Babbie, E. 2001. The practice of social research, Oxford, Oxford University Press.
Barker, T., Ekins, P. & Foxon, T. 2007. The macro-economic rebound effect and the UK
economy. Energy Policy, 35, 4935-4946.
275
Bibliography
Bates, S. R. 2006. Making time for change: On temporal conceptualizations within (critical
realist) approaches to the relationship between structure and agency. Sociology, 40,
143.
BAULINKS. 2004. Neue KfW-Förderprogramme ab Januar 2005 [Online]. Available:
http://m.baulinks.de/webplugin/2004/1573.php4 [Accessed 08/08 2010].
BAULINKS. 2009. Änderungen bei KfW-Förderprogrammen [Online]. Available:
http://www.baulinks.de/webplugin/2009/0539.php4 [Accessed 01/05 2012].
BAULINKS.
2010.
Kommentare
zur
Mittelkürzung
beim
CO2Gebäudesanierungsprogramm
[Online].
Available:
http://www.baulinks.de/webplugin/2010/0992.php4 [Accessed 20/03 2012].
Baumgartner, F. R. 2006. Punctuated Equilibrium Theory and Environmental Policy. In:
Repetto, R. C. (ed.) Punctuated equilibrium and the dynamics of U.S.
environmental policy. New Haven ; London: Yale University Press.
Baumgartner, F. R. 2009. Lobbying and policy change : who wins, who loses, and why,
Chicago ; London, University of Chicago Press.
Baumgartner, F. R. 2011. Ideas and policy change. Governance Symposium on Policy
Paradigms and Social Learning. Suffolk University, Boston.
Baumgartner, F. R., Breunig, C., Green-Pedersen, C., Jones, B. D., Mortensen, P. B.,
Nuytemans, M. & Walgrave, S. 2009. Punctuated Equilibrium in Comparative
Perspective. American Journal of Political Science, 53, 603.
Baumgartner, F. R., Green-Pedersen, C. & Jones, B. D. 2006. Comparative studies of
policy agendas. Journal of European Public Policy, 13, 959-974.
Baumgartner, F. R. & Jones, B. D. 1991. Agenda Dynamics and Policy Subsystems. The
Journal of Politics, 53, 1044-1074.
Baumgartner, F. R. & Jones, B. D. 1993. Agendas and instability in American politics,
Chicago ; London, University of Chicago Press.
Baumgartner, F. R. & Jones, B. D. 2002a. Policy dynamics, Chicago, London, University
of Chicago Press.
Baumgartner, F. R. & Jones, B. D. 2002b. Positive and Negative Feedback in Politics. In:
Baumgartner, F. R. & Jones, B. D. (eds.) Policy dynamics. Chicago: University of
Chicago Press.
Baumgartner, F. R. & Jones, B. D. 2009. Agendas and instability in American politics,
Chicago ; London, The University of Chicago Press.
BEEF. 2002. Economic Instruments to improve household energy [Online]. Available:
http://web.archive.org/web/20071006140613/http://www.ukace.org/pubs/consult/b
eefTreasResp.pdf [Accessed 05/01 2013].
Béland, D. 2010. Reconsidering Policy Feedback. Administration & Society, 42, 568-590.
Bennett, A. 1912. An Arnold Bennett calendar: A quotation from Arnold Bennett for every
day in the year. Compiled by Frank Bennett, New York, George Doran Co.
Bennett, A. 2004a. Case Study Methods: Design, Use, and Comparative Advantages. In:
Sprinz, D. F. & Wolinsky-Nahmias, Y. (eds.) Models, numbers, and cases :
methods for studying international relations. Ann Arbor: University of Michigan
Press.
Bennett, A. 2004b. Case Study Methods: Design, Use, and Comparative Advantages. In:
Sprinz, D. F. & Wolinsky-Nahmias, Y. (eds.) Models, Numbers, and Cases:
Methods for Studying International Relations. Ann Arbor: University of Michigan
Press.
Bennett, A. & Elman, C. 2006. Qualitative research: Recent Developments in Case Study
Methods. Annual Review of Political Science, 9, 455-476.
276
Bibliography
Bennett, G. 1991. Air pollution control in the European Community : implementation of
the EC Directives in the twelve member states, London, Graham & Trotman.
Bertoldi, P. & Rezessy, S. 2008 ACEEE Summer Study on Energy Efficiency, 2008. 2233.
Bhaskar, R. 1978. A realist theory of science, Hassocks, Harvester Press.
Birkland, T. A. 1997. After disaster : agenda setting, public policy, and focusing events,
Washington, DC, Georgetown University Press.
Blaikie, N. W. H. 1993. Approaches to social enquiry, Cambridge, Polity.
Bloomberg. 2012. Spot Carbon Dioxide (CO2) Emissions EUA Price/Europe [Online].
Available: http://www.bloomberg.com/quote/EUETSSY1:IND [Accessed 22/03
2012].
Blumstein, C. 2010. Program evaluation and incentives for administrators of energyefficiency programs: Can evaluation solve the principal/agent problem? Energy
Policy, 38, 6232-6239.
BMU 2000. Nationales Klimaschutzprogramm. Umwelt, 11, I-XVI.
BMU. 2003. Gemeinsame Pressemitteilung BMVBW / BMU / KfW. Bundesregierung stellt
zusätzlich jährlich 160 Millionen Euro für Sanierung im Gebäudebereich bereit.
Fördermöglichkeiten im KfW-Programm werden ausgeweitet [Online]. Available:
http://www.bmu.de/pressearchiv/15_legislaturperiode/pm/pdf/2111.pdf [Accessed
17/08 2010].
BMU 2005. Nationales Klimaschutzprogramm 2005. Berlin: BMU.
BMU 2006. Fourth National Report by the Government of the Federal Republic of
Germany (4th National Communication). Berlin: BMU.
BMU. 2007. Das Integrierte Energie- und Klimaprogramm der Bundesregierung [Online].
Available:
http://www.bmu.de/files/pdfs/allgemein/application/pdf/hintergrund_meseberg.pdf
[Accessed 06/05 2010].
BMU.
2010.
Kyoto
Protocol
[Online].
Available:
http://www.bmu.de/english/climate/international_climate_policy/kyoto_protocol/d
oc/41823.php [Accessed 05/05 2010].
BMVBS 2010. Wohnen und Bauen in Zahlen 2009/2010. Berlin, Bonn: BMVBS.
BMVBS 2011. Wohnen und Bauen in Zahlen. 2010/2011. Berlin: BMVBS.
BMWI & BMF. 2008. Beschäftigungssicherung durch Wachstumsstärkung.
Maßnahmenpaket
der
Bundesregierung
[Online].
Available:
http://www.bmwi.de/BMWi/Redaktion/PDF/W/wachstumspaket-breg-november08,property=pdf,bereich=bmwi,sprache=de,rwb=true.pdf [Accessed 17/08 2010].
BMWi & BMU 2007. Bericht zur Umsetzung der in der Kabinettsklausur am
23./24.08.2007 in Meseberg beschlossenen Eckpunkte für ein Integriertes Energieund Klimaprogramm. Berlin: BMU and BMWi.
BMWi & BMU 2010. Energiekonzept für eine umweltschonende, zuverlässige und
bezahlbare Energieversorgung. Endfassung. 28. September 2010. Berlin: BMWi.
Boardman, B. 1991a. Fuel poverty : from cold homes to affordable warmth, London,
Belhaven.
Boardman, B. 1991b. Fuel poverty is different. Policy Studies, 12, 30-41.
Boardman, B. 1993. Opportunities and constraints posed by fuel poverty on policies to
reduce the greenhouse effect in britain. Applied Energy, 44, 185-195.
Boardman, B. 1995. Freezing out the poor. New Economy, 2, 152-156.
Boardman, B. 2004. New directions for household energy efficiency: evidence from the
UK. Energy Policy, 32, 1921-1933.
277
Bibliography
Boardman, B. 2007. Examining the carbon agenda via the 40% House scenario. Building
Research & Information, 35, 363-378.
Boardman, B. 2010. Fixing fuel poverty : challenges and solutions, London, Earthscan.
Boardman, B. 2012. Achieving zero. Oxford: Environmental Change Institute, University
of Oxford.
Bonoli, G. 2010. The Political Economy of Active Labor-Market Policy. Politics &
Society, 38, 435-457.
Boonekamp, P. G. & Eichhammer, W. 2007. Applied policy measures for energy savings
in end-use sectors in EU countries: what is best? Proceedings of the ECEEE
Summer Study 2007, 263-273.
Boushey, G. 2012. Punctuated Equilibrium Theory and the Diffusion of Innovations.
Policy Studies Journal, 40, 127-146.
Bowen, G. A. 2009. Document analysis as a qualitative research method. Qualitative
Research Journal, 9, 27-40.
Bower, J. 2003. Why OFGEM? Oxford: Oxford Institute for Energy Studies.
Boykoff, M. 2011. Media coverage of climate change / global warming [Online].
Available:
http://sciencepolicy.colorado.edu/media_coverage/uk/index.html
[Accessed 12/12 2011].
Boykoff, M. & Rajan, S. R. 2007. Signals and noise. EMBO Rep, 8, 207-211.
BPIE 2011. Europe's Buildings under the Microscope. A country-by-country review of the
energy performance of buildings. Brussels: BPIE.
Bradshaw, J. 1983. Introduction. In: Bradshaw, J. & Harris, T. (eds.) Energy and social
policy. London: Routledge and Kegan.
Brady, D. W. 1988. Critical elections and congressional policy making, Stanford, Calif,
Stanford University Press.
BRE 2006. Research into the effectiveness of loft insulation. Garston: BRE.
Breunig, C. & Koski, C. 2006. Punctuated Equilibria and Budgets in the American States.
Policy Studies Journal, 34, 363-379.
Brookes, L. 1990. The greenhouse effect: the fallacies in the energy efficiency solution.
Energy Policy, 18, 199-201.
Brown, I. 1990. Least Cost Planning in the Gas Industry, London, OFGAS.
BSI. 2010. BSI: Sparpläne der Bundesregierung verheerend für den Klimaschutz und die
Entwicklung der Städte – „Kürzungen im investiven Bereich sind unsinnig und
wirtschaftlich falsch“ [Online]. Available: http://www.bsi-web.de/?p=164
[Accessed 20/03 2012].
Bundesregierung 2000. Nationales Klimaschutzprogramm. Umwelt, 11, I-XVI.
Bundesregierung 2003. Jahresbericht der Bundesregierung zum Stand der deutschen
Einheit. Berlin: Bundesregierung.
Bundesregierung. 2005a. 20-Punkte-Programm: Zur Stärkung von Konjunktur und
Wachstum
[Online].
Available:
http://archiv.bundesregierung.de/bpaexport/artikel/12/815312/multi.htm [Accessed
12/08 2010].
Bundesregierung. 2005b. Regierungserklärung von Bundeskanzler Gerhard Schröder vor
dem Deutschen Bundestag am 17. März. 17.03.2005 [Online]. Available:
http://archiv.bundesregierung.de/bpaexport/regierungserklaerung/73/803473/multi.
htm [Accessed 17/08 2010].
Bundesregierung. 2006. Politik im Dreiklang Investieren - Sanieren - Reformieren: Das
Kabinett hat sich auf der Klausurtagung in Genshagen vom 9. bis 10. Januar 2006
auf ein Investitionsprogramm von 25 Milliarden Euro geeinigt [Online]. Available:
278
Bibliography
http://archiv.bundesregierung.de/Content/DE/Archiv16/Artikel/2006/01/2006-0110-ergebnispapier-der-klausurtagung-in-genshagen-material-fuer-die-presse.html
[Accessed 17/08 2010].
Bundesvereinigung Bauwirtschaft. 2003. Robl: Bundesregierung ist mit Förderung der
Altbausanierung auf dem richtigen Weg. [Online]. Available: http://www.bvbauwirtschaft.de/zdb.nsf/0/52A2F55713227BF5C125718E002FCA41?openDocum
ent [Accessed 08/08 2010].
Bundesvereinigung Bauwirtschaft. 2006. CO2-Gebäudesanierungsprogramm: Das Bauund Ausbaugewerbe begrüßt Vereinfachungen und Zuschussvariante [Online].
Available:
http://www.bvbauwirtschaft.de/zdb.nsf/0/4BFE92D605A06A06C12572880037EF6E?openDocu
ment [Accessed 08/08 2010].
Bundesvereinigung Bauwirtschaft. 2010. Robl: Mittel für energetische Gebäudesanierung
verstetigen
[Online].
Available:
http://www.bvbauwirtschaft.de/bvb.nsf/0/142544E861CE3CEBC12576D5005016AA [Accessed
08/08 2010].
Bürger, V., Rohde, C., Eichhammer, W. & Schlomann, B. 2012. Energieeinsparquote für
Deutschland? Bewertung des Instruments der Energieeinsparquote (Weiße
Zertifikate) auf seine Eignung als Klimaschutzinstrument für Deutschland.
Freiburg: Öko-Institut.
Burnham, W. D. 1970. Critical elections and the mainsprings of American politics, New
York, Norton.
Butterwegge, C. 2012. Krise und Zukunft des Sozialstaates, Wiesbaden, VS Verlag für
Sozialwissenschaften.
Cairney, P.
7. A ‘Multiple Lenses’ Approach to Policy Change: The Case of Tobacco
Policy in the UK. British Politics, 2, 45-68.
Cairney, P. A. 2009. Multiple Lenses and Multiple Narratives in Public Policy: Are They
Complementary or Contradictory? SSRN eLibrary.
CAMPACT. 2010. Wiesehügel:"Bundesregierung gefährdet Klimaschutzziele und
Tausende Arbeitsplätze"/ Kritik an Kürzung von Wärmedämmungsprogramm
60.000 Protest-Emails an Ramsauer und Schäuble [Online]. Available:
http://www.presseportal.de/pm/64126/1564301/wiesehuegel-bundesregierunggefaehrdet-klimaschutzziele-und-tausende-arbeitsplaetze-kritik-an [Accessed 01/05
2012].
Campbell, A. L. 2002a. Self-interest, social security, and the distinctive participation
patterns of senior citizens. The American Political Science Review, 96, 565.
Campbell, D. T. 1975. "Degrees of freedom" and the case study. Comparative Political
Studies, 8, 178-193.
Campbell, J. L. 2002b. Ideas, Politics, and Public Policy. Annual Review of Sociology, 28,
21-38.
Campbell, J. L. 2004. Institutional change and globalization, Princeton, N.J. ; Oxford,
Princeton University Press.
Capano, G. 2009. Understanding Policy Change as an Epistemological and Theoretical
Problem. Journal of Comparative Policy Analysis: Research and Practice, 11, 731.
Cashore, B. & Howlett, M. 2007. Punctuating Which Equilibrium? Understanding
Thermostatic Policy Dynamics in Pacific Northwest Forestry. American Journal of
Political Science, 51, 532-551.
279
Bibliography
Cayre, E., Allibe, B., Laurent, M.-H. & Osso, D. 2011. There are people in the house!
How the results of purely technical analysis of residential energy consumption are
misleading for energy policies. ECEEE 2011 Summer Study: Energy efficiency
first: The foundation of a low-carbon society. Belambra Presqu'île de Giens,
France.
CDU, CSU & FDP 2009. Wachstum. Bildung. Zusammenhalt. Koalitionsvertrag zwischen
CDU, CSU und FDP. 17. Legislaturperiode. Berlin: CDU, CSU, FDP.
CDU, CSU & SPD 2005. Gemeinsam für Deutschland. Mit Mut und Menschlichkeit.
Koalitionsvertrag von CDU, CSU und SPD. Rheinbach: Union-Betriebs GmbH.
Centre for Economics and Business Research 2011. British Gas Home Energy Report
2011. An assessment of the drivers of domestic natural gas consumption. London:
Centre for Economics and Business Research.
Chabal, P. M. 2003. Do Ministers Matter? The Individual Style of Ministers in
Programmed Policy Change. International Review of Administrative Sciences, 69,
29-49.
Chitnis, M., Sorrell, S., Druckman, A., Firth, S. & Jackson, T. 2012. Estimating direct and
indirect rebound effects for UK households. SLRG Working Paper 01-12. Surrey:
Sustainable Lifestyles Research Group.
Chung, H. & Thewissen, S. 2011. Falling Back on Old Habits? A Comparison of the
Social and Unemployment Crisis Reactive Policy Strategies in Germany, the UK
and Sweden. Social Policy & Administration, 45, 354-370.
Clausnitzer, K. D., Fette, M. & Gabriel, J. 2011. Evaluation der KfW-Programme „KfWKommunalkredit - Energetische eb udesanierung“, „Energieeffizient Sanieren –
Kommunen“ und „Sozial investieren – Energetische Gebäude-sanierung“ der Jahre
2007 bis 2010. Bremen: Bremer Energie Institut.
Clausnitzer, K. D., Fette, M., Gabriel, J., Diefenbach, N., Loga, T. & Wosniok, W. 2010.
Effekte der Förderfälle des Jahres 2009 des CO2-Gebäudesanierungsprogramms
und des Programms „Energieeffizient Sanieren“. Bremen: Bremer Energie Institut.
Clausnitzer, K. D., Gabriel, J., Diefenbach, N., Loga, T. & Wosniok, W. 2007. Ermittlung
von Effekten des KfW-CO2-Gebäudesanierungsprogramms. Entwicklung der
Methodik und Ergebnisse der Berichtsperioden 2005 und 2006. Bremen: Bremer
Energie Institut.
Clausnitzer, K. D., Gabriel, J., Diefenbach, N., Loga, T. & Wosniok, W. 2008. Effekte des
CO2-Gebäudesanierungsprogramms 2007. Bremen: Bremer Energie Institut.
Clausnitzer, K. D., Gabriel, J., Diefenbach, N., Loga, T. & Wosniok, W. 2009. Effekte des
CO2-Gebäudesanierungsprogramms 2008. Bremen: Bremer Energie Institut.
Clegg, D. 2010. Labour market policy in the crisis: the UK in comparative perspective.
The journal of poverty and social justice, 18, 5-17.
Climate Policy Initiative. 2011. Steueranreize zur F rderung energetischer Sanierungen
[Online].
Available:
http://climatepolicyinitiative.org/wpcontent/uploads/2011/12/Steueranreize-zur-Förderung-energetischerSanierungen.pdf [Accessed 20/07 2012].
Clinton, J., Geller, H. & Hirst, E. 1986. Review of Government and Utility Energy
Conservation Programs. Annual review of energy, 11, 95-142.
Cohen, M. D., March, J. & Olsen, J. 1972. A garbage can model of organizational choice.
Administrative Science Quarterly, 17, 1.
Collier, D. 1993. The comparative method. Political Science: The State of the Discipline
II, 105-119.
280
Bibliography
Collins, E. C. 1992. Qualitative research as art: Toward a holistic process. Theory into
Practice, 31, 181.
Cook, J. 2007. The Book of Positive Quotations, Minneapolis, Fairview Press.
Cortell, A. P. & Peterson, S. 1999. Altered States: Explaining Domestic Institutional
Change. British Journal of Political Science, 29, 177-203.
Cruickshank, J. 2003. Critical realism : the difference it makes, London, Routledge.
Dahl, R. A. & Lindblom, C. E. 1953. Politics, economics and welfare : planning and
politico-economic systems resolved into basic social processes, New York, Harper.
Dawkins, L. A. 1987. The politics of energy conservation and industrial interests in
Britain. Parliamentary Affairs, 40, 250-264.
DECC 2009a. Extending the Carbon Emissions Reduction Target. Consultation on a
CERT framework for the period April 2011 to December 2012. London: DECC.
DECC 2009b. Sustainable Energy Report 2009. Reporting progress against English
household sector energy efficiency aims. London: DECC.
DECC 2009c. The UK Low Carbon Transition Plan. London: DECC.
DECC 2010a. Estimated impacts of energy and climate change policies on energy prices
and bills. London: DECC.
DECC 2010b. Extending the Carbon Emissions Reduction Target to December 2012.
Impact Assessment. London: DECC.
DECC. 2010c. Insulation for the nation – 3.5 million more homes to be lagged [Online].
Available: http://www.decc.gov.uk/en/content/cms/news/pn10_075/pn10_075.aspx
[Accessed 07/07 2010].
DECC. 2010d. Retail prices index UK: fuel components in the UK/relative to GDP
deflator
(QEP
2.1.1
and
2.1.2)
[Online].
Available:
http://www.decc.gov.uk/en/content/cms/statistics/source/prices/prices.aspx
[Accessed 29/06/2010 2010].
DECC. 2010e. Social and Environmental Guidance to the Gas and Electricity Markets
Authority
[Online].
Available:
http://www.ofgem.gov.uk/About%20us/Documents1/file37517.pdf
[Accessed
07/07 2010].
DECC 2011a. Annual Report on Fuel Poverty Statistics 2011. London: DECC.
DECC 2011b. Control Framework for DECC levy funded spending. Questions and
answers. London: DECC.
DECC 2011c. Digest of United Kingdom Energy Statistics 2011. London: DECC.
DECC 2011d. The Green Deal and Energy Company Obligation Impact Assessment.
London: DECC.
DECC 2011e. National Energy Efficiency Data framework. Report on the development of
the data-framework and initial analysis. London: DECC.
DECC 2011f. Quarterly energy prices. September 2011. London: DECC.
DECC 1 a. 1 uidelines to Defra / DECC’s H . Conversion Factors for Company
Reporting: Methodology Paper for Emission Factors. London: DECC.
DECC 2012b. Annual report on fuel poverty statistics 2012. London: DECC.
DECC 2012c. Digest of United Kingdom energy statistics 2012: long-term trends.
London: DECC.
DECC 2012d. Energy consumption in the United Kingdom. London: DECC.
DECC 2012e. National Energy Efficiency Data-Framework. Summary of analysis using
the National Energy Efficiency Data-Framework. London: DECC.
DEFRA 2001. Energy Efficiency Commitment 2002-2005. Consultation Proposals.
London: DEFRA.
281
Bibliography
DEFRA 2004a. The energy efficiency commitment from April 2005 - consultation
proposals. London: DEFRA.
DEFRA 2004b. Energy Efficiency: The Government’s Plan for Action. London: DEFRA.
DEFRA 2006a. The Energy Efficiency Commitment April 2008 to March 2011: Initial
consultation. London: DEFRA.
DEFRA 2006b. Synthesis of Climate Change Policy Evaluations. London: DEFRA.
DEFRA 2007a. Carbon Emissions Reduction Target. April 2008 to March 2011.
Consultation Proposals. May 2007. London: DEFRA.
DEFRA 2007b. UK Energy Efficiency Action Plan 2007. London: DEFRA.
DEFRA 2008. Explanatory memorandum to the Electricity and Gas (Carbon Emissions
Reduction) Order 2008. London: DEFRA.
Denzin, N. K. 1970. The research act in sociology: a theoretical introduction to
sociological methods, London, Butterworths.
DETR 1998. UK climate change programme : consultation paper. London: TSO.
DETR 2000. Climate Change. Draft UK Programme. London: DETR.
Deutscher Bundestag. 1993a. Antwort der Bundesregierung auf die Große Anfrage
"Umsetzung der Empfehlungen der Enquete Kommission "Vorsorge zum Schutz der
Erdatmosphäre" durch die Bundesregierung. Durcksache 12/4280 [Online]. Bon:
Deutscher
Bundestag.
Available:
http://dipbt.bundestag.de/dip21/btd/12/085/1208557.pdf [Accessed 14/02 2012].
Deutscher Bundestag. 1993b. Beschlußempfehlung und Bericht des Außschusses für
Umwelt, Naturschutz und Reaktorsicherheit (17. Außschuß) zu dem Dritten Bericht
der Enquete-Kommission "Vorsorge zum Schutz der Erdatmosphäre". Drucksache
12/1136
[Online].
Bonn:
Deutscher
Bundestag.
Available:
http://dip21.bundestag.de/dip21/btd/12/011/1201136.pdf.
Deutscher Bundestag. 1994. Beschluß der Bundesregierung zur Verminderung der CO2Emissionen und anderer Treibhausgasemissionen in der Bundesrepublik
Deutschland auf der Grundlage des Dritten Berichts der Interministeriellen
Arbeitsgruppe "CO2-Reduktion" (IMA "CO2-Reduktion). Drucksache 12/8557
[Online].
Bonn:
Deutscher
Bundestag.
Available:
http://dipbt.bundestag.de/dip21/btd/12/085/1208557.pdf [Accessed 14/02 2012].
Deutscher Bundestag. 1997a. Antwort der Bundesregierung auf die Kleine Anfrage des
Abgeordneten Klaus-Jürgen Warnick und der Gruppe der PDS - Drucksache
13/8186 - Zukunft des Wohnraum-Modernisierungsprogramms der Kreditanstalt
für
Wiederaufbau.
Drucksache
13/8303
[Online].
Available:
http://dipbt.bundestag.de/dip21/btd/13/083/1308303.asc [Accessed 08/08 2010].
Deutscher Bundestag. 1997b. Beschluß der Bundesregierung zum Klimaschutzprogramm
der Bundesrepublik Deutschland auf der Basis des Vierten Berichts der
Interministeriellen Arbeitsgruppe "CO2-Reduktion" (IMA "CO2-Reduktion).
Drucksache 13/8936 [Online]. Bonn: Deutscher Bundestag. Available:
http://dip21.bundestag.de/dip21/btd/13/089/1308936.pdf [Accessed 14/02 2012].
Deutscher Bundestag. 2000. Unterrichtung durch die Bundesregierung. Nationales
Klimaschutzprogramm. Fünfter Bericht der Interministeriellen Arbeitsgruppe
„CO2-Reduktion“. Drucksache 14/4729 [Online]. Bonn: Deutscher Bundestag.
Available: http://dipbt.bundestag.de/dip21/btd/14/047/1404729.pdf [Accessed
14/02 2012].
Deutscher Bundestag. 2002. Antwort der Bundesregierung auf die Kleine Anfrage der
Abgeordneten Dr. Martina Krogmann, Matthias Wissmann, Ernst Hinsken,
weiterer Abgeordneter und der Fraktion der CDU/CSU. Drucksache 14/9280
282
Bibliography
[Online]. Berlin. Available: http://dipbt.bundestag.de/doc/btd/14/095/1409514.pdf
[Accessed 14/02 2012].
Deutscher Bundestag. 2004. Unterrichtung durch die Bundesregierung. Jahresbericht der
Bundesregierung zum Stand der Deutschen Einheit 2004. Drucksache 15/3796
[Online].
Available:
http://dipbt.bundestag.de/dip21/btd/15/037/1503796.pdf
[Accessed 08/08 2010].
Deutscher Bundestag. 2007. Unterrichtung durch die Bundesregierung Bericht der
Bundesregierung über die Entwicklung der Finanzhilfen des Bundes und der
Steuervergünstigungen für die Jahre 2005 bis 2008 (21. Subventionsbericht).
Drucksache
16/6275
[Online].
Available:
http://dipbt.bundestag.de/dip21/btd/16/062/1606275.pdf [Accessed 08/08 2010].
Deutscher Bundestag. 2010a. Antwort der Bundesregierung auf die Kleine Anfrage der
Abgeordneten Daniela Wagner, Bettina Herlitzius, Ingrid Nestle, weiterer
Abgeordneter und der Fraktion BÜNDNIS 90/DIE GRÜNEN. Drucksache 17/2995
[Online].
Berlin:
Deutscher
Bundestag.
Available:
http://dip21.bundestag.de/dip21/btd/17/033/1703341.pdf [Accessed 15/02 2012].
Deutscher Bundestag. 2010b. Antwort der Bundesregierung auf die Kleine Anfrage der
Abgeordneten Daniela Wagner, Bettina Herlitzius, Ingrid Nestle, weiterer
Abgeordneter und der Fraktion BÜNDNIS 90/DIE GRÜNEN. Drucksache 17/2995
[Online].
Available:
http://dip21.bundestag.de/dip21/btd/17/033/1703341.pdf
[Accessed 15/12 2010].
Deutscher Bundestag. 2010c. Drucksache 17/2346. Antrag der Abgeordneten Michael
Groß, Sören Bartol, Uwe Beckmeyer, Martin Burkert, Bernhard Brinkmann
(Hildesheim), Petra Ernstberger, Iris Gleicke, Ulrike Gottschalck, Hans-Joachim
Hacker, Gustav Herzog, Johannes Kahrs, Ute Kumpf, Kirsten Lühmann, Thomas
Oppermann, Florian Pronold, Dr. Frank-Walter Steinmeier und der Fraktion der
SPD. Angekündigte Mittelkürzung beim CO2-Gebäudesanierungsprogramm
zurücknehmen
[Online].
Available:
http://dipbt.bundestag.de/dip21/btd/17/023/1702346.pdf [Accessed 08/08 2010].
Deutscher Bundestag. 2010d. Drucksache 17/2395. Antrag der Abgeordneten Stephan
Kühn, Daniela Wagner, Bettina Herlitzius, Ingrid Nestle, Winfried Hermann, Dr.
Anton Hofreiter, Dr. Valerie Wilms, Cornelia Behm, Hans-Josef Fell, Kai Gehring,
Ulrike Höfken, Bärbel Höhn, Sylvia Kotting-Uhl, Oliver Krischer, Undine Kurth
(Quedlinburg), Nicole Maisch, Friedrich Ostendorff, Dr. Hermann Ott, Dorothea
Steiner, Markus Tressel und der Fraktion BÜNDNIS 90/DIE GRÜNEN. CO2Gebäudesanierungsprogramm fortführen - Mit energetischer Sanierung
Konjunktur ankurbeln, Arbeitsplätze sichern und Klima schützen [Online].
Available: http://dipbt.bundestag.de/dip21/btd/17/023/1702395.pdf [Accessed
08/08 2010].
Deutscher Bundestag. 2010e. Kleine Anfrage der Abgeordneten Daniela Wagner, Bettina
Herlitzius, Ingrid Nestle, weiterer Abgeordneter und der Fraktion BÜNDNIS
90/DIE
GRÜNEN.
Drucksache
17/2995
[Online].
Available:
http://dip21.bundestag.de/dip21/btd/17/029/1702995.pdf [Accessed 15/12 2010].
Deutscher Bundestag. 2011a. Antwort der Bundesregierung auf die Kleine Anfrage der
Abgeordneten Michael Groß, Uwe Beckmeyer, Sören Bartol, weiterer
Abgeordneter und der Fraktion der SPD. Drucksache 17/5546 [Online]. Available:
http://dipbt.bundestag.de/dip21/btd/17/058/1705817.pdf [Accessed 06/03 2012].
Deutscher Bundestag. 2011b. Gesetzentwurf der Fraktionen der CDU/CSU und FDP.
Entwurf eines Gesetzes zur Änderung des Gesetzes zur Errichtung eines
283
Bibliography
Sonderverm gens „Energie- und Klimafonds“ (EKFG-ÄndG). Drucksache
17/6075
[Online].
Available:
http://dipbt.bundestag.de/dip21/btd/17/060/1706075.pdf [Accessed 06/03 2012].
Deutscher Bundestag. 2012. Kleine Anfrage der Abgeordneten Daniela Wagner, SvenChristian Kindler, Bettina Herlitzius, Ingrid Nestle, Lisa Paus, Harald Ebner, Dr.
Anton Hofreiter, Stephan Kühn, Dr. Valerie Wilms und der Fraktion BÜNDNIS
90/DIE GRÜNEN. Entwicklung des CO2-Gebäudesanierungsprogramms nach
Halbierung der Einnahmen im Energie- und Klimafonds und hälftiger Zuweisung
der Mittel des Energie- und Klimafonds. Drucksache 17/8523 [Online]. Available:
http://dipbt.bundestag.de/dip21/btd/17/085/1708523.pdf [Accessed 06/03 2012].
Umweltminister Gabriel weist Kritik an Klimaschutz-Projekten zurück, 2007. Directed by
Deutschlandradio Kultur. Germany.
Dey, I. 1993. Qualitative data analysis : a user-friendly guide for social scientists,
London, Routledge.
Diefenbach, N., Enseling, A., Loga, T., Hertle, H., Jahn, D. & Duscha, M. 2005. Beiträge
der EnEV und des KfW-CO2-Gebäudesanierungsprogramms zum Nationalen
Klimaschutzprogramm. Darmstadt: Institut Wohnen und Umwelt GmbH.
Diefenbach, N., Loga, T., Gabriel, J. & Fette, M. 2011. Monitoring der KfW-Programme
„Energieeffizient Sanieren“ 1 und „Ökologisch / Energieeffizient Bauen“
6
- 2010. Bremen: Bremer Energie Institut.
Diefenbach, T. 2009. Are case studies more than sophisticated storytelling?:
Methodological problems of qualitative empirical research mainly based on semistructured interviews. Quality & Quantity, 43, 875-894.
Dietrich, S. 2009. Ausgeburt der Krise [Online]. Frankfurter Allgemeine Zeitung.
Available: http://www.faz.net/aktuell/politik/foederalismusreform-ausgeburt-derkrise-1923398.html [Accessed 02/02 2012].
DIW & ISI 1997. Energie-Effizienz-Indikatoren: Statistische Grundlagen, theoretische
Fundierung und Orientierungsbasis für die politische Praxis.
DoE 1992. Climate change: Our national programme for CO2 emissions. London:
Department of the Environment.
Doll, C., Eichhammer, W., Fleiter, T., Ragwitz, M., Schade, W., Schleich, J., Schlomann,
B., Sensfuss, F., Walz, R., Wietschel, M., Harthan, R. O., Matthes, F. C., Hansen,
P., Kleemann, M., Markewitz, P., Martinsen, D., Ziesing, H.-J. & Jakob, M. 2008.
Wirtschaftlicher Nutzen des Klimaschutzes. Kostenbetrachtung ausgewählter
Einzelmaßnahmen der Meseberger Beschlüsse zum Klimaschutz. Karlsruhe,
Berlin, Jülich, Zürich: Fraunhofer ISI, Forschungszentrum Jülich, Öko-Institut,
ETH Zürich.
Dowding, K. 2001. There Must Be End to Confusion: Policy Networks, Intellectual
Fatigue, and the Need for Political Science Methods Courses in British
Universities. Political Studies, 49, 89-105.
DPG 1985. Warnung vor einer drohenden Klimakatastrophe. Bonn: DPG.
Dresner, S. & Ekins, P. 2006. Economic Instruments to Improve UK Home Energy
Efficiency without Negative Social Impacts. Fiscal Studies, 27, 47-74.
DTI 1998a. A fair deal for consumers : modernising the framework for utility regulation :
the response to consultation. London: TSO.
DTI 1998b. A Fair Deal for Consumers: Modernising the Framework of Utility
Regulation. London: TSO.
284
Bibliography
DTI 1999. A fair deal for consumers : modernising the framework for utility regulation,
the future of gas and electricity regulation, the government's proposals for
legislation. London: TSO.
DTI 2001. The UK Fuel Poverty Strategy. London: DTI.
DTI 2003. Energy White Paper: Our energy future - creating a low carbon economy.
London: TSO.
DTI 2007. Meeting the Energy Challenge. A White Paper on Energy. London: TSO.
Eadie, A. 1992. Special Report on the Electricity Industry: Cleanliness through efficiency:
Alison Eadie examines how companies are responding to the pressure to reduce
energy consumption. The Independent. 08/07/1992 ed.
EEFA 2010. EEFA-Energiemodell. Münster: EEFA.
EEPH 2010. Energy Efficiency Partnership for Homes. Annual Review 2009-2010.
London: EEPH.
Eichhammer, W., Schlomann, B. & Kling, N. 2006. Energy Efficiency Policies and
Measures in Germany 2006. Karlsruhe: Fraunhofer Institute for Systems and
Innovation Research.
Ekins, P. 2004. Step changes for decarbonising the energy system: research needs for
renewables, energy efficiency and nuclear power. Energy Policy, 32, 1891-1904.
Ekins, P. & Lees, E. 2008. The impact of EU policies on energy use in and the evolution of
the UK built environment. Energy Policy, 36, 4580-4583.
Ekins, P. & Lockwood, M. 2011. Tackling fuel poverty during the transition to a lowcarbon economy. York: Joseph Rowntree Foundation.
Ekins, P., Russell, A. & Hargreaves, C. 2002. Reducing carbon emissions through
improved household energy efficiency in the UK. Journal of environmental policy
& planning, 4, 41-65.
Eldredge, N. & Gould, S. 1972. Punctuated equilibria: An alternative to phyletic
gradualism. Models in paleobiology. San Francisco: Cooper and Co.
Element Energy Ltd, Quantum Strategy & Technology, 2008. Assessment of the size of
the UK household energy efficiency market. London: EEPH.
Elo, S. & Kyngäs, H. 2008. The qualitative content analysis process. Journal of Advanced
Nursing, 62, 107-115.
ENBAUSa. 2010a. Dena will 5 Milliarden für Sanierungsprogramm der KfW [Online].
Available:
http://www.enbausa.de/finanzierung-beratung/aktuelles/artikel/denawill-5-milliarden-fuer-sanierungsprogramm-der-kfw-954.html [Accessed 01/05
2012].
EnBauSa. 2010b. KfW-Gelder für Sanierung sind wieder frei [Online]. Available:
http://www.enbausa.de/finanzierung-beratung/aktuelles/artikel/kfw-gelder-fuersanierung-sind-wieder-frei-925.html [Accessed 08/08 2010].
EnBauSa. 2010c. KfW stoppt Zusagen für Sanierungsprogramme [Online]. Available:
http://www.enbausa.de/finanzierung-beratung/aktuelles/artikel/kfw-stopptzusagen-fuer-sanierungsprogramme-847.html [Accessed 08/08 2010].
EnBauSa. 2010d. Kürzung bei der Gebäudesanierung stößt auf harte Kritik [Online].
Available: http://www.enbausa.de/daemmung-fassade/aktuelles/artikel/kuerzungbei-der-gebaeudesanierung-stoesst-auf-harte-kritik-1310.html [Accessed 08/08
2010].
ENDS 1994a. Holes open up in CO2 strategy. ENDS Report, 232.
ENDS 1994b. Ofgas "drives coach and horses" through CO2 strategy. ENDS Report, 231.
ENDS 1995. CO2 strategy crumbles as Ofgas blocks energy saving schemes. ENDS
Report, 241.
285
Bibliography
ENDS 1998. DETR rules out levy to fund domestic energy efficiency drive. ENDS Report,
286.
ENDS 1999a. Energy Saving Trust demands urgent action by Government and regulator.
ENDS Report, 288.
ENDS 1999b. MPs call for "energy hierarchy" as Ofgem shifts on energy efficiency.
ENDS Report, 294.
ENDS 2005. Energy supply companies ahead of the game on efficiency targets. ENDS
Report, 367.
ENDS 2006a. Carbon reduction targets proposed for EEC3. ENDS Report, 379, 37-38.
ENDS 2006b. DEFRA lists measures for third phase of EEC. ENDS Report. 05/10/2006
ed.
ENDS 2006c. Domestic energy suppliers face carbon reduction targets. ENDS Report.
03/08/2006 ed.
ENDS 2006d. Suppliers on track to meet energy efficiency targets early. ENDS Report,
379, 12.
ENDS 2008. Government favours energy efficiency over windfall tax. ENDS Report, 404,
13-14.
Energy Monitoring Company 2008. Disaggregation of the Energy Savings Achieved from
Insulation in EESoP 3 and the Energy Efficiency Commitment. London: EST.
Energy Saving Trust 1994. Recommendations on the standards of performance in energy
efficiency for the regional electricity companies. London: OFFER.
Energy Saving Trust 2004. Monitoring Energy Savings achieved from Insulation Measures
installed in Gas Heated Homes in SoP3 and EEC Schemes. London: Energy Saving
Trust.
Engels, A., Knoll, L. & Huth, M. 2008. Preparing for the ‘real’market: national patterns of
institutional learning and company behaviour in the European emissions trading
scheme (EU ETS). European environment, 18, 276.
Enquete Kommission 1988. Erster Zwischenbericht der Enquete-Kommission „Vorsorge
zum Schutz der Erdatmosphäre", Bonn, Bonner Universitäts-Buchdruckerei.
Enquete Kommission 1990. Dritter Bericht der Enquete Kommission Vorsorge zum Schutz
der Erdatmosphäre zum Thema Schutz der Erde, Bonn, Deutscher Bundestag.
Enquete Kommission 1994. Schlußbericht der Enquete-Kommission "Schutz der
Erdatmosphäre" des Deutschen Bundestages. Drucksache 12/8600, Bonn,
Deutscher Bundestag.
ERA. 2004a. The Energy Efficiency Commitment from April 2005 - Consultation
proposals. Response from the Energy Retail Association [Online]. Available:
http://www.energy-uk.org.uk/publication/finish/35/395.html [Accessed 20/01
2012].
ERA 2004b. Energy suppliers views on the feasability of delivering an EEC 2 obligation
and likely costs post 2005. London: ERA.
ERA. 2006. Our Energy Challenge – The DTI Energy Review. Submission from the
Energy Retail Association [Online]. ERA. Available: http://docs.energyuk.org.uk/publication/finish/33-2006/384-era-submission-to-the-energy-review2006.html [Accessed 21/01 2012].
ERA. 2007. ERA response to Carbon Emissions Reduction Target April 2008 - March
2011 [Online]. ERA. Available: http://docs.energy-uk.org.uk/publication/finish/272007/359-era-response-to-carbon-emissions-reduction-target-march-2007.html
[Accessed 20/01 2012].
286
Bibliography
Erhon, H. 2007. Bedarf – Verbrauch: Ein Reizthema ohne Ende oder die Chance für
sachliche Energieberatung? [Online]. Fraunhofer-Institut für Bauphysik.
Available: http://www.buildup.eu/publications/1810 [Accessed 06/09 2012].
Erlandson, D. A. 1993. Doing naturalistic inquiry : a guide to methods, Newbury Park,
Calif. ; London, Sage.
Esping-Andersen, G. 1990. The three worlds of welfare capitalism, Cambridge, Polity.
Eto, J., Kito, S., Shown, L. & Sonnenblick, R. 1995. Where Did the Money Go? The Cost
and Performance of the Largest Commercial Sector DSM Programs. Berkeley:
Lawrence Berkeley National Laboratory.
Eyre, N. 1997. Barriers to energy efficiency more than just market failure. Energy and
Environment, 8, 25-43.
Eyre, N. Regulation of energy suppliers to save energy – lessons from the UK debate.
British Institute of Energy Economics Academic Conference, 2008 Oxford. BIEE.
Eyre, N. & Staniaszek, D. 2005. Energy efficiency in the UK Energy White Paper – How
did it get a central role? European Council for an Energy Efficienct Economy
Summer Study. Mandelieu.
Farquharson, K. 2005. A Different Kind of Snowball: Identifying Key Policymakers.
International Journal of Social Research Methodology, 8, 345 - 353.
Fels, M. F. & Keating, K. M. 1993. Savings from Demand-Side Management Programs in
US Electric Utilities. Annual Review of Energy and the Environment, 18, 57-88.
Feyerabend, P. 1981. Problems of empiricism, Cambridge, Cambridge University Press.
Fischer, C., Parry, I. W. H. & Pizer, W. A. 2003. Instrument choice for environmental
protection when technological innovation is endogenous. Journal of Environmental
Economics and Management, 45, 523-545.
Fleischer, T. 1997. CO2-Minderung in Deutschland: Nationale Aufgaben und
Handlungsfelder. In: Kopfmüller, J. & Coenen, R. (eds.) Risiko Klima. Der
Treibhauseffekt als Herausforderung für Wissenschaft und Politik. Frankfurt:
Campus.
Focus. 2008. SPD will „nicht kleckern“ – CDU für Steuerentlastungen bereits in 2009 ...
Konjunkturprogramm: SPD will „nicht kleckern“ – CDU für Steuerentlastungen
bereits
in
2009
[Online].
Available:
http://www.focus.de/politik/deutschland/konjunkturprogramm-spd-will-nichtkleckern-cdu-fuer-steuerentlastungen-bereits-in-2009_aid_344379.html [Accessed
20/02 2012].
Forfori, F. 2006. Evaluation of the British Energy Efficiency Commitment within the
framework of the AID-EE project. Ecofys, Wuppertal Institute, Lund University.
Forum Ökologisch-Soziale Marktwirtschaft. 2010. Gebäudesanierung ausweiten Finanzierung
über
Heizstoffsteuer
[Online].
Available:
http://www.foes.de/pdf/2010-09-PM-Gebaeudesanierung.pdf [Accessed 01/05
2012].
FPAG 2003. Fuel Poverty Advisory Group for England: First Annual Report - 2002/2003
London: DTI.
FPAG 2004. Fuel Poverty Advisory Group for England: Second Annual Report 2003/2004. London: DTI.
FPAG 2005. Fuel Poverty Advisory Group for England: Third Annual Report - 2004/2005.
London: DTI.
FPAG 2006. Fuel Poverty Advisory Group for England: Fourth Annual Report - 2005.
London: DTI.
287
Bibliography
FPAG 2007. Fuel Poverty Advisory Group for England: Fourth Annual Report - 2006.
London: DTI.
FPAG 2008. Fuel Poverty Advisory Group for England: Sixth Annual Report - 2007.
London: FPAG.
FPAG 2009. Fuel Poverty Advisory Group for England: Seventh Annual Report - 2008.
London: FPAG.
FPAG 2010. Fuel Poverty Advisory Group for England: Eighth Annual Report - 2009.
London: FPAG.
FPAG 2011. Fuel Poverty Advisory Group for England: Ninth Annual Report - 2010.
London: FPAG.
Frankfurter Allgemeine Zeitung 2007. Bundesregierung entschärft das Klima-Programm.
Frankfurter Allgemeine Zeitung, 188, 13.
Frankfurter Allgemeine Zeitung. 2008a. „Die Lage ist noch schlechter geworden“
[Online]. Available: http://www.faz.net/aktuell/wirtschaft/wirtschaftspolitik/imgespraech-peer-steinbrueck-die-lage-ist-noch-schlechter-geworden-1714841.html
[Accessed 20/02 2012].
Frankfurter Allgemeine Zeitung. 2008b. Politik will Konjunktur ankurbeln [Online].
Available: http://www.faz.net/aktuell/wirtschaft/konjunktur/bundesbank-erwartetstagnation-politik-will-konjunktur-ankurbeln-1712951.html
[Accessed
20/02
2012].
Freeman, G. P. 1985. National Styles and Policy Sectors: Explaining Structured Variation.
Journal of Public Policy, 5, 467-496.
Friedrich, M. 2006. Evaluation des Modernisierungsratgebers. Endbereicht f r die
co online gemeinn tzige mbH. Berlin: co online
Frondel, M. 2008. Energiesparen: Warum wird Beratung gefördert? Zeitschrift für
Energiewirtschaft, 32, 97.
Frondel, M., Ritter, N., Schmidt, C. M. & Vance, C. 2010. Economic impacts from the
promotion of renewable energy technologies: The German experience. Energy
Policy, 38, 4048-4056.
Frondel, M. & Schmidt, C. M. 2005. Evaluating environmental programs: The perspective
of modern evaluation research. Ecological Economics, 55, 515-526.
Fudge, S., Peters, M., Mulugetta, Y. & Jackson, T. 2011. Paradigms, Policy and
Governance: The Politics of Energy Regulation in the UK Post-2000.
Environmental Policy and Governance, 21, 291-302.
Gabriel, J. & Balmert, D. 2007. Effekte des CO2-Gebäudesanierungsprogramms 2005 und
2006. Zusatzauswertung Dezember 2007. Bremen: Bremer Energie Institut.
Galvin, R. 2010. Thermal upgrades of existing homes in Germany: The building code,
subsidies, and economic efficiency. Energy and Buildings, 42, 834–844.
GdW. 2010. Kürzung von Städtebauförderung und KfW-Mitteln zur Gebäudesanierung
angekündigt
[Online].
Available:
http://web.gdw.de/pressecenter/pressearchiv/2010/187-qramsauerskuerzungsplaene-haben-verheerende-auswirkungen-auf-die-kuenftige-qualitaetdes-wohnens-sowie-auf-deutschlands-staedteq [Accessed 20/03 2012].
GEA 2012. Global Energy Assessment - Toward a Sustainable Future, Cambridge
University Press, Cambridge, UK and New York, NY, USA and the International
Institute for Applied Systems Analysis, Laxenburg, Austria.
Geller, H. S. & Attali, S. 2005. The Experience with Energy Efficiency Policies and
Programmes in IEA Countries: Learning from the Critics. Paris: IEA.
288
Bibliography
George, A. L. & Bennett, A. 2005. Case studies and theory development in the social
sciences, Cambridge, Mass., MIT Press.
George, A. L. & McKeown, T. J. 1985. Case Studies and Theories of Organizational
Decision Making. Advances in Information Processing in Organizations, 2, 21-58.
Gillingham, K., Newell, R. & Palmer, K. 2006. Energy efficiency policies: A retrospective
examination.
Givel, M. 2006. Punctuated Equilibrium in Limbo: The Tobacco Lobby and U.S. State
Policymaking from 1990 to 2003. Policy Studies Journal, 34, 405-418.
Givel, M. 2008. Assessing Material and Symbolic Variations in Punctuated Equilibrium
and Public Policy Output Patterns. Review of Policy Research, 25, 547-561.
Givel, M. 2010. The Evolution of the Theoretical Foundations of Punctuated Equilibrium
Theory in Public Policy. The review of policy research, 27, 187.
Givel, M. 2012. Nonpunctuated and Sweeping Policy Change: Bhutan Tobacco Policy
Making from 1991 to 2009. Review of Policy Research, 29, 645-660.
l ser, J.
Laudel, .
. Experteninterviews und qualitative Inhaltsanalyse,
Wiesbaden, VS Verlag f r Sozialwissenschaften.
Goldstone, J. A. 1997. Methodological Issues in Comparative Macrosociology.
Comparative Social Research, 16, 107-120.
Goodman, L. A. 1961. Snowball Sampling. The Annals of Mathematical Statistics, 32,
148-170.
Grasl, M. & König, M. 2010. Von Außen getrieben. Die Finanzpolitik der Großen
Koalition 2005-2009. In: Egle, C. & Zohlnhöfer, R. (eds.) Die zweite Große
Koalition : eine Bilanz der Regierung Merkel 2005-2009. 1. Aufl. ed. Wiesbaden:
VS Verlag für Sozialwissenschaften.
Greening, L. A., Greene, D. L. & Difiglio, C. 2000. Energy efficiency and consumption the rebound effect - a survey. Energy Policy, 28, 389-401.
Greller, M., Schröder, F., Hundt, V., Mundry, B. & Papert, O. 2010. Universelle
Energiekennzahlen für Deutschland – Teil 2: Verbrauchskennzahlentwicklung nach
Baualtersklassen. Bauphysik, 32, 1-6.
Grice, A. & Russel, B. 2008. Economic woes force Chancellor to delay green taxes. The
Independent. 11/03/2008 ed.
Grösche, P. & Vance, C. 2009. Willingness to Pay for Energy Conservation and FreeRidership on Subsidization: Evidence from Germany. Energy Journal, 30, 135153.
Haas, R. & Biermayr, P. 2000. The rebound effect for space heating: Empirical evidence
from Austria. Energy Policy, 28, 403-410.
Hacker, J. S. 2002. The divided welfare state : the battle over public and private social
benefits in the United States, Cambridge, Cambridge University Press.
Hall, P. A. 1983. Policy Innovation and the Structure of the State: The PoliticsAdministration Nexus in France and Britain. The Annals of the American Academy
of Political and Social Science, 466, 43-59.
Hall, P. A. 1986. Governing the economy : the politics of state intervention in Britain and
France, Cambridge, Polity.
Hall, P. A. 1993. Policy paradigms, social learning, and the state: the case of economic
policy making in britain. Comparative Politics, 25, 275-296.
Hall, P. A. 2003. Aligning ontology and methodology in comparative research. In:
Mahoney, J. & Rueschemeyer, D. (eds.) Comparative historical analysis in the
social sciences. Cambridge: Cambridge University Press.
289
Bibliography
Hall, P. A. 2008. Systematic Process Analysis: when and how to use it. European Political
Science: EPS, 7, 304.
Hall, P. A. & Soskice, D. W. 2001. Varieties of capitalism : the institutional foundations of
comparative advantage, Oxford, Oxford University Press.
Hansard. 2008. House of Commons Debates 10 January 2008. Volume No. 470. Part No.
29
[Online].
Available:
http://www.publications.parliament.uk/pa/cm200708/cmhansrd/cm080110/debtext/
80110-0005.htm [Accessed 01/05 2012].
Hansen, D. 2007. Crisis and Perspectives on Policy Change, Vällingby, Swedish National
Defence College, Crismart.
Hay, C. 2006. Political Ontology. In: Goodin, R. E. & Tilly, C. (eds.) The Oxford
Handbook of Contextual Political Analysis. Oxford: Oxford University Press.
Hay, C. 2007. Does ontology trump epistemology? Notes on the directional dependence of
ontology and epistemology in political analysis. Politics, 27, 115.
Heclo, H. 1976. Conclusion: policy dynamics. In: Rose, R. (ed.) The Dynamics of public
policy : a comparative analysis. London: Sage.
Helm, D. 2002. Energy policy: security of supply, sustainability and competition. Energy
Policy, 30, 173.
Helm, D. 2003. Energy, the state, and the market : British energy policy since 1979,
Oxford, Oxford University Press.
Henderson, G., Staniaszek, D., Anderson, B. & Phillipson, M. Energy savings from
insulation improvements in electrically heated dwellings in the UK. ECEE 2003
Summer Study, 2003. 325-334.
Hens, H., Parijs, W. & Deurinck, M. 2010. Energy consumption for heating and rebound
effects. Energy and Buildings, 42, 105-110.
Hesse, H. 1972. The Glass Bead Game (Magister Ludi), Middlesex, Penguin.
Hewett, M. J., Dunsworth, T. S., Miller, T. A. & Koehler, M. J. 1986. Measured versus
predicted savings from single retofits: a sample study. Energy and Buildings, 9, 6573.
Hills, J. 2011. Fuel Poverty: The problem and its measurement. London: DECC.
Hills, J. 2012. Getting the measure of fuel poverty. Final Report of the Fuel Poverty
Review. London: DECC.
Hirst, E. 1990. Progress and Potential in Evaluating Energy Efficiency Programs.
Evaluation Review, 14, 192-205.
Hirst, E. & Hannon, B. 1979. Effects of energy conservation in residential and commercial
buildings. Science, 205, 656-661.
Hirst, E., Trimble, J., Goeltz, R. & Cardell, S. 1983. Methods to Deal with Self-Selection
in Estimating Conservation Program Energy Savings: Use of Synthetic Data to
Test Alternative Approaches. Tennessee: Oak Ridge National Laboratory.
HM
Government.
1986.
1986
Gas
Act
[Online].
Available:
http://www.opsi.gov.uk/RevisedStatutes/Acts/ukpga/1986/cukpga_19860044_en_1
[Accessed 07/07 2010].
HM
Government.
1989.
Electricity
Act
1989
[Online].
Available:
http://www.opsi.gov.uk/acts/acts1989/ukpga_19890029_en_4#pt1-pb8-l1g41
[Accessed 07/07 2010].
HM Government 1990. This Common Inheritance: Britain's Environmental Strategy.
London: HMSO.
HM Government 1992. This Common Inheritance: The Second Year Report, London,
HMSO.
290
Bibliography
HM Government 1994. Climate Change: the UK Programme. United Kingdom's Report
under the Framework Convention on Climate Change (Cm 2427). London: HMSO.
HM Government 2006. Climate Change The UK Programme 2006. London: TSO.
HM Government. 2008. Home Energy Saving Programme [Online]. Available:
http://www.number10.gov.uk/wp-content/uploads/energy-savingprogramme110908.pdf [Accessed 01/07 2010].
HM Treasury 2011. Control framework for DECC levy-funded spending. London: HM
Treasury.
Holling, C. S. 1973. Resilience and Stability of Ecological Systems. Annual Review of
Ecology and Systematics, 4, 1-23.
Hong, S. H., Oreszczyn, T. & Ridley, I. 2006. The Impact of energy efficient
refurbishment on the space heating fuel consumption in English dwellings. Energy
and Buildings, 38, 1171-1181.
House of Commons Business and Enterprise Committee 2008. Energy prices, fuel poverty
and Ofgem. Eleventh Report of Session 2007–08. London: TSO.
House of Commons Energy Committee 1991a. Energy Efficiency. Third Report. Session
1990-91. Volume I. Report together with the proceedings of the Committee.
London: HMSO.
House of Commons Energy Committee 1991b. Energy Efficiency. Third Report. Session
1990-91. Volume II. Memoranda of Evidence. London: HMSO.
House of Commons Energy Committee 1991c. Energy Efficiency. Third Report. Session
1990-91. Volume III. Minutes of Evidence. London: HMSO.
House of Commons Environment Committee 1993a. Energy efficiency in buildings.
Session 1992-93. Volume I. Report together with the proceedings of the
Committee relating to the report. London: HMSO.
House of Commons Environment Committee 1993b. Energy efficiency in buildings.
Session 1992-93. Volume II. Minutes of evidence. London: HMSO.
House of Commons Environment Committee 1994. Session 1993-94. Energy Efficiency:
the Role of OFGAS. London: HMSO.
House of Commons Environmental Audit Select Committee 1999. Seventh Report. Energy
Efficiency. London: House of Commons.
House of Commons Information Office 2008. Statutory Instruments. Factsheet L7.
London: House of Commons.
House of Commons Trade and Industry Committee 1994. The Work of Ofgas. Session
1993-94. Minutes of Evidence, 25/01/1994. London: HMSO.
Howlett, M. 1997. Issue-Attention and Punctuated Equilibria Models Reconsidered: An
Empirical Examination of the Dynamics of Agenda-Setting in Canada. Canadian
Journal of Political Science, 30, 3.
Howlett, M. 2007. Re-visiting the new orthodoxy of policy dynamics: the dependent
variable and re-aggregation problems in the study of policy change. Canadian
political science review, 1, 50.
Howlett, M. & Cashore, B. 2009. The Dependent Variable Problem in the Study of Policy
Change: Understanding Policy Change as a Methodological Problem. Journal of
Comparative Policy Analysis: Research and Practice, 11, 33 - 46.
Howlett, M. & Migone, A. 2011. Charles Lindblom is alive and well and living in
punctuated equilibrium land. Policy and Society, 30, 53-62.
Hussain, A. 2006. Third rise in energy prices is on its way. The Sunday Times, 11/09/2006,
p.7.
291
Bibliography
Hutchinson, P., Read, R. J. & Sharrock, W. W. 2008. There is no such thing as a social
science : in defence of Peter Winch, Aldershot ; Burlington, Vt., Ashgate.
IEA. 2006. Energy Policies of IEA Countries - the United Kingdom [Online]. Available:
http://www.iea.org/publications/free_new_Desc.asp?PUBS_ID=1860
[Accessed
05/07 2012].
IEA. 2007. IEA commends UK energy policy for supporting investment and energy
efficiency, but outlines challenges on future security of supply [Online]. Available:
http://www.iea.org/press/pressdetail.asp?PRESS_REL_ID=217 [Accessed 25/07
2012].
IER & PROGNOS 2004. Analyse der Wirksamkeit von CO2-Minderungsmaßnahmen im
Energiebereich und ihre Weiterentwicklung. Basel: Prognos.
Ifo Institute. 2012. Long time-series of the Business Climate for German trade and
industry and its two components, the business situation and the expectations
[Online]. Available: http://www.cesifo-group.de/portal/page/portal/_alt_ifocvnill/awinfo/d1index/10indexgsk [Accessed 05/02 2012].
IPCC 1995. Climate Change 1995: A report of the Intergovernmental Panel on Climate
Change. Second Assessment Report of the Intergovernmental Panel on Climate
Change, Cambridge, Cambridge University Press.
IPCC 2007. Climate change 2007. Mitigation of climate change : contribution of Working
Group III to the fourth assessment report of the Intergovernmental Panel on
Climate Change, Cambridge, Cambridge University Press.
Jameson, A. 2008. Hit us with a tax and the public will pay for it, energy chiefs say
[Online].
Available:
http://www.thetimes.co.uk/tto/business/industries/utilities/article2197636.ece
[Accessed 30/06 2010].
Jensen, C. 2009. Policy Punctuations in Mature Welfare States. Journal of Public Policy,
29, 287-303.
Jensen, C. 2011. Focusing events, policy dictators and the dynamics of reform. Policy
Studies, 32, 143-158.
Jevons, S. 1865. The Coal Question - Can Britain Survive?
John, P. 1998. Analysing public policy, London, Pinter.
John, P. 2003. Is there life after policy streams, advocacy coalitions, and punctuations:
using evolutionary theory to explain policy change? Policy Studies Journal, 31,
481.
John, P. 2006. The policy agendas project: a review. Journal of European Public Policy,
13, 975.
John, P. & Jennings, W. 2010. Punctuations and Turning Points in British Politics: The
Policy Agenda of the Queen?s Speech, 1940?2005. British Journal of Political
Science, First View, 1-26.
Jones, A. 2001. Social Responsibility and the Utilities. Journal of Business Ethics, 34,
219-229.
Jones, B. D. & Baumgartner, F. R. 2005. The Politics of attention: how government
prioritizes problems, Chicago, University of Chicago Press.
Jones, B. D. & Baumgartner, F. R. 2012. From There to Here: Punctuated Equilibrium to
the General Punctuation Thesis to a Theory of Government Information
Processing. Policy Studies Journal, 40, 1-20.
Jones, B. D., Baumgartner, F. R., Breunig, C., Wlezien, C., Soroka, S., Foucault, M.,
François, A., Green-Pedersen, C., Koski, C., John, P., Mortensen, P. B., Varone, F.
292
Bibliography
& Walgrave, S. 2009. A General Empirical Law of Public Budgets: A Comparative
Analysis. American Journal of Political Science, 53, 855-873.
Jones, B. D., Sulkin, T. & Larsen, H. A. 2003. Policy Punctuations in American Political
Institutions. The American Political Science Review, 97, 151-169.
Jordan, A.
. The United Kingdom: From Policy ‘Taking’ to Policy ‘Shaping’. In:
Jordan, A. & Liefferink, D. (eds.) Environmental Policy in Europe: The
Europeanization of National Environmental Policy. London: Routledge.
Joskow, P. L. 1993. What does a negawatt really cost? Further thoughts and evidence. The
Electricity journal, 6, 14.
Kaßner, R., Wilkens, M., Wenzel, W. & Ortjohan, J. 2010. Online-Monitoring zur
Sicherstellung energetischer Zielwerte in der Baupraxis, Paper for 3. Effizienz
Tagung Bauen + Modernisieren, 19-20 November 2010, Hannover. [Online].
Available:
http://www.energy-check.de/wpcontent/uploads/2010/11/EFT_2010_ortjohann_2010-10-18.pdf [Accessed 06/09
2012].
Katznelson, I. 1998. The Doleful Dance of Politics and Policy: Can Historical
Institutionalism Make a Difference? The American Political Science Review, 92,
191-197.
Kelly, S. 2011. Do homes that are more energy efficient consume less energy?: A
structural equation model for England’s residential sector. EPRG Working Paper
1117
Cambridge Working Paper in Economics 1139. Cambridge.
Kelly, S., Crawford-Brown, D. & Pollitt, M. G. 2012. Building performance evaluation
and certification in the UK: Is SAP fit for purpose? Renewable and Sustainable
Energy Reviews, 16, 6861-6878.
Kennedy, S. & Pagnamenta, R. 2008. Fuel firms face threat of windfall profits tax. The
Times. 01/08/2008 ed.
Key, V. O., Jr. 1955. A Theory of Critical Elections. The Journal of Politics, 17, 3-18.
KfW.
2003.
KfW-CO2-Gebäudesanierungsprogramm
[Online].
Available:
http://www.rbbaunatal.de/downloads/mgebsan.pdf [Accessed 08/08 2010].
KfW. 2006a. Ab sofort deutlich verbesserte Kreditkonditionen für energetische
Sanierungsmaßnahmen
an
Wohngebäuden
[Online].
Available:
http://www.kfw.de/DE_Home/Presse/Pressearchiv/2006/20060202.jsp [Accessed
15/08 2010].
KfW. 2006b. Brief an die mit der KfW in Verbindung stehenden Geschäftspartner
[Online]. Available: http://www.enev-check.de/KfW_01.Feb.2006.pdf [Accessed
08/08 2010].
KfW. 2006c. CO2-Gebäudesanierungsprogramm erweitert: Neben der verbesserten
Darlehensförderung startet am 1.1.2007 eine Zuschussvariante [Online].
Available: http://www.kfw.de/DE_Home/Presse/Pressearchiv/2006/20061127.jsp
[Accessed 17/08 2010].
KfW. 2006d. CO2-Gebäudesanierungsprogramm. Stand 09/2006 [Online]. Available:
http://mudrich.de/pdf/CO2.pdf [Accessed 20/03 2012].
KfW. 2007. CO2-Gebäudesanierungsprogramm - Kredit. Stand 01/2007 [Online].
Available:
http://www.delta-q.de/export/sites/default/de/downloads/7-81_kfw_co2-gebaeudesanierung_130_kredit_2007_01_04.pdf [Accessed 20/03
2012].
KfW. 2008. KfW Förderbank: Ausweitung der Programme für Energieeffizientes Bauen
und
Sanieren
ab
Januar
2009
[Online].
Available:
293
Bibliography
http://www.kfw.de/DE_Home/Presse/Pressearchiv/PDF/2008/085_D_Energieeffizi
entes_Bauen_und_Sanieren.pdf [Accessed 07/08 2010].
KfW. 2009. Neue Förderprogramme für Energieeffizienz und Klimaschutz Tiefensee:
Energieeffizient Bauen wird noch leichter gemacht [Online]. Available:
http://www.kfw.de/DE_Home/Presse/Pressekonferenzen/PDFDokumente_2009/09-0319_PE_D_Neue_Foerderprogr._f._Energieeff._u._Klimasch.pdf [Accessed 15/08
2010].
King, A. 1985. Book review: Agendas, Alternatives, and Public Policy. Journal of Public
Policy, 5, 281-283.
King, G., Keohane, R. O. & Verba, S. 1994. Designing social inquiry : scientific inference
in qualitative research, Princeton ; Chichester, Princeton University Press.
Kingdon, J. W. 1984. Agendas, alternatives, and public policies, Boston, Little, Brown.
Kingdon, J. W. 1995. Agendas, alternatives, and public policies, New York, Longman.
Kingdon, J. W. 2002. The reality of public policy making. In: Danis, M., Clancy, C. M. &
Churchill, L. R. (eds.) Ethical dimensions of health policy. Oxford: Oxford
University Press.
Kleemann, M. & Hansen, P. 2005. Evaluierung der CO2-Minderungsmaßnahmen im
Gebäudebereich. Jülich: Forschungszentrum Jülich.
Kleemann, M., Heckler, R., Kraft, A. & Kuckshinrichs, W. 2003. Klimaschutz und
Beschäftigung durch das KfW-Programm zur CO2-Minderung und das KfW-CO2Gebäudesanierungsprogramm. Jülich: Forschungszentrum Jülich GmbH.
Kline, M. 1990. Mathematical thought from ancient to modern times, Oxford, Oxford
University Press.
Kluve, J. 2010. The effectiveness of European active labor market programs. Labour
Economics, 17, 904-918.
Koelble, T. A. 1995. The New Institutionalism in Political Science and Sociology.
Comparative Politics, 27, 231-243.
Kolbe, R. H. & Burnett, M. S. 1991. Content-Analysis Research: An Examination of
Applications with Directives for Improving Research Reliability and Objectivity.
The Journal of Consumer Research, 18, 243-250.
Kopatz, M., Spitzer, M. & Christanell, A. 2010. Energiearmut. Stand der Forschung,
nationale Programme und regionale Modellprojekte in Deutschland, Österreich und
Großbritannien. Wuppertal Papers. Wuppertal: Wuppertal Institut f r Klima,
Umwelt, Energie GmbH.
Kramer, M., Allyn, B. J., Blight, J. G. & Welch, D. A. 1990. Remembering the Cuban
Missile Crisis: Should We Swallow Oral History? International Security, 15, 212218.
Kreitler, V. 1991. On customer choice and free ridership in utility programs. 5th Int.
Energy Program Eval. Conf. Chicaco.
Krippendorff, K. 1969. Models of messages: three prototypes. In: Gerbner, G., Holsti, O.
R., Krippendorff, K., Paisley, W. & Stone, P. J. (eds.) The analysis of
communication content: developments in scientific theories and computer
techniques. New York, Chichester: Wiley.
Krippendorff, K. 2004. Content analysis : an introduction to its methodology, Thousand
Oaks, London, SAGE.
Krück, C., Borchers, J. & Weingart, P. 1999. Climate research and climate politics in
Germany: Assets and hazards of consensus-based risk management. In: Miller, C.
294
Bibliography
& Edwards, P. (eds.) Changing the atmosphere: science and the politics of global
warming. Cambridge, MA: MIT Press.
Kuckshinrichs, W., Hansen, P.
Kronenberg, T.
9. esamtwirtschaftliche CO Vermeidungskosten der energetischen
eb udesanierung und Kosten der
F rderung
f r
den
Bundeshaushalt
im
Rahmen
des
CO eb udesanierungsprogramms.
J lich:
Institut
f r
Energieforschung
Systemforschung
und
Technologische
Entwicklung
(IEF-STE).
Forschungszentrum J lich.
Kuckshinrichs, W., Kronenberg, T. & Hansen, P. 2010. The social return on investment in
the energy efficiency of buildings in Germany. Energy Policy, 38, 4317-4329.
Kvale, S. 2007. Doing interviews, Los Angeles ; London, SAGE.
Labour Party. 1997. New Labour: Because Britain Deserves Better [Online]. Available:
http://www.labour-party.org.uk/manifestos/1997/1997-labour-manifesto.shtml
[Accessed 15/01 2013].
Landman, T. 2008. Issues and methods in comparative politics : an introduction, London,
Routledge.
Lasswell, H. D. 1971. A pre-view of policy sciences, New York, American Elsevier Pub.
Co.
Lauber, V. & Mez, L. 2004. Three Decades of Renewable Electricity Policies in Germany.
Energy & Environment, 15, 599-623.
Lawrence, G. 2000. Who Regulates the Regulators? CRI Occasional Paper. Bath: CRI.
Lawson, N. 1982. Speech given at the Fourth Annual International Conference,
International Association of Energy Economists. Churchill College, Cambridge, 28
June 1982.
LE Energy Ltd & SRC International 1992. Demand Side Measures. Report to the Office of
Electricity Regulation. London: LE Energy Ltd.
Lees, C. 2007a. Environmental Policy in the United Kingdom and Germany. German
Politics, 16, 164-183.
Lees, E. 2006. Evaluation of the Energy Efficiency Commitment 2002-05. Wantage: Eoin
Lees Energy.
Lees, E. 2007b. European Experience of White Certificates. World Energy Council,
ADEME.
Lees, E. 2008. Evaluation of the Energy Efficiency Commitment 2005-08. Wantage: Eoin
Lees Energy.
LeLoup, L. T. & Moreland, W. B. 1978. Agency Strategies and Executive Review: The
Hidden Politics of Budgeting. Public Administration Review, 38, 232-239.
Levine, M. D., Koomey, J. G., McMahon, J. E., Sanstad, A. H. & Hirst, E. 1995. Energy
efficiency policy and market failures. Annual Review of Energy and the
Environment, 20, 535-555.
LGA 2008. Switch off, switched on. London: LGA.
Lieberman, R. C. 2002. Ideas, institutions, and political order: Explaining political change.
American Political Science Review, 96, 697-712.
Lijphart, A. 1971. Comparative politics and the comparative method. The American
Political Science Review, 65, 682-693.
Lindblom, C. E. 1959. The Science of "Muddling Through". Public Administration
Review, 19, 79-88.
Lindblom, C. E. 1979. Still Muddling, Not Yet Through. Public Administration Review,
39, 517-526.
295
Bibliography
Loga, T. & Diefenbach, N. 2005. Entwicklung eines vereinfachten, statistisch
abgesichterten Verfahrens zur Erhebung von Gebäudedaten für die Erstellung des
Energieprofils von Gebäuden. Darmstadt: Institut Wohnen und Umwelt.
Loga, T., Diefenbach, N. & Born, R. 2011. Deutsche Gebäudetypologie. Beispielhafte
Maßnahmen zur Verbesserung der Energieeffizienz von typischen Wohngebäuden.
Darmstadt: Institute Wohnen und Umwelt.
Lorenz, A. 2010. Neue Muster der Verfassungspolitik unter der Großen Koalition. In:
Egle, C. & Zohlnhöfer, R. (eds.) Die zweite Große Koalition: eine Bilanz der
Regierung Merkel 2005-2009. 1. Aufl. ed. Wiesbaden: VS Verlag für
Sozialwissenschaften.
Loughran, D. & Kulick, J. 2004. Demand-side management and energy efficiency in the
United States. Energy Journal, 25, 19-41.
Lovins, A. B. 1976. Energy Strategy: The Road Not Taken. Foreign Affairs, 55, 65-96.
Lovins, A. B. 1977. Soft energy paths: toward a durable peace, Harmondsworth, Penguin.
Lovins, A. B. 1985. Saving gigabucks with negawatts. Public Utilities Fortnightly, 115,
19-26.
Lowe, R. 2009. Policy and strategy challenges for climate change and building stocks.
Building Research & Information, 37, 206-212.
Lowi, T. J. 1972. Four Systems of Policy, Politics, and Choice. Public administration
review, 32, 298-310.
Lowry, W. 2006. Potential Focusing Projects and Policy Change. Policy Studies Journal,
34, 313-335.
Madlener, R. & Hauertmann, M. 2011. Rebound effects in German residential heating: Do
ownership and income matter? Aachen: Institute for Future Energy Consumer
Needs and Behavior (FCN), RWTH Aachen.
Mahoney, J. 1999. Nominal, Ordinal, and Narrative Appraisal in Macrocausal Analysis.
The American Journal of Sociology, 104, 1154-1196.
Mahoney, J. 2000a. Strategies of Causal Inference in Small-iV Analysis. Sociological
Methods & Research, 28, 387.
Mahoney, J. 2000b. Strategies of Causal Inference in Small-N Analysis. Sociological
Methods & Research, 28, 387.
Mahoney, J. 2007. Qualitative methodology and comparative politics. Comparative
Political Studies, 40, 122.
Malm, E. 1996. An actions-based estimate of the free rider fraction in electric utility DSM
programs. The Energy Journal, 17, 41.
March, J. G. & Olsen, J. P. 1984. The New Institutionalism: Organizational Factors in
Political Life. The American Political Science Review, 78, 734-749.
March, J. G. & Olsen, J. P. 1989. Rediscovering institutions : the organizational basis of
politics, New York, Oxford, Free Press, Maxwell Macmillan.
Marsh, D. 2008. Historical Institutionalism and Critical Realism: Necessary Bedfellows?
[Online].
Available:
http://www.arts.usyd.edu.au/arts/government_international_relations/uploads/Histo
ricalInstitutionalismandCriticalRealism.pdf [Accessed 08/08 2010].
Marsh, D. & Smith, M. J. 2001. There is more than one way to do political science: On
different ways to study policy networks. Political Studies, 49, 528-541.
Martin, C. & Watson, M. 2006. Measurement of energy savings and comfort levels in
houses receiving insulation upgrades. London: EST.
Maso, I. 1989. The necessity of being flexible. Quality & Quantity, 23, 161-170.
296
Bibliography
Mayring, P. 2000. Qualitative Content Analysis. Forum Qualitative Sozialforschung /
Forum: Qualitative Social Research [Online], 1. Available: http://nbnresolving.de/urn:nbn:de:0114-fqs000220 [Accessed 08/10/2010].
McAnulla, S. 2005. Making hay with actualism? The need for a realist concept of
structure. Politics, 25, 31.
McKinsey. 2007. Appendix: UK Abement cost curve. London:CBI.
Meijerink, S. 2005. Understanding policy stability and change. the interplay of advocacy
coalitions and epistemic communities, windows of opportunity, and Dutch coastal
flooding policy 1945-20031. Journal of European Public Policy, 12, 1060.
Mettler, S. & Soss, J. 2004. The consequences of public policy for democratic citizenship:
Bridging policy studies and mass politics. Perspectives on Politics, 2, 55.
Meuser, M. & Nagel, U. 1991. Experteninterviews – vielfach erprobt, wenig bedacht.
Qualitative Sozialforschung: Konzepte, Methoden, Analysen. Opladen:
Westdeutscher Verlag.
Michaelowa, A. 2005. The German wind energy lobby: how to promote costly
technological change successfully. European Environment, 15, 192-199.
Mieterbund. 2010. Verbände und Gewerkschaft: Energetische Gebäudemodernisierung
unterfinanziert
–
Klimapolitik
unglaubwürdig
[Online].
Available:
http://www.mieterbund.de/pressemitteilung.html?&no_cache=1&tx_ttnews[tt_new
s]=274&cHash=3efd050ac6fc87e5f1f6b4f2106d82d0 [Accessed 01/05 2012].
Mill, J. S. 1843. A system of logic, ratiocinative and inductive : being a connected view of
the principles of evidence and the methods of scientific investigation, London, John
W. Parker.
Milne, G. & Boardman, B. 2000. Making cold homes warmer: the effect of energy
efficiency improvements in low-income homes A report to the Energy Action
Grants Agency Charitable Trust. Energy Policy, 28, 411-424.
Mintrom, M. 1996. Advocacy coalitions, policy entrepreneurs, and policy change. Policy
Studies Journal, 24, 420.
Mitchell, C. 2009. The political economy of sustainable energy, Basingstoke, Palgrave
Macmillan.
Mitchell, C., Bauknecht, D. & Connor, P. M. 2006. Effectiveness through risk reduction: a
comparison of the renewable obligation in England and Wales and the feed-in
system in Germany. Energy Policy, 34, 297-305.
Mucciaroni, G. 1992. The Garbage Can Model & the Study of Policy Making: A Critique.
Polity, 24, 459-482.
Murphy, L., Meijer, F. & Visscher, H. 2012. Effective National Energy Performance
Instruments for Existing Dwellings? Lessons from Front-Runners. Proceedings of
Retrofit 2012 Conference, Salford Quays 24-26 January.
Nadel, S. & Keating, K. M. 1991. Engineeringestimates versus impact evaluation results:
how do they compare and why? 1991 International Energy Program Evaluation
Conference.
National Audit Office 1998. Improving Energy Efficiency Financed by a Charge on
Customers. London: National Audit Office.
Naughton, P. 2008. Gordon Brown facing embarrassment over free household energy
claim. The Times. 11/09/2008 ed.
NEA. 2000. Basic Principles of new-style EESoPs [Online]. NEA. Available:
http://www.nea.org.uk/consultation-paper-responses/results?selectyear=2000&formController=consultation-paper297
Bibliography
responses%2F&executeForm=SearchForm&action_results=
[Accessed
25/01
2012].
NEA. 2001. Energy Efficiency Commitment 2002-2005 Consultation Proposals [Online].
Available:
http://www.nea.org.uk/consultation-paper-responses/results?selectyear=2001&formController=consultation-paperresponses%2F&executeForm=SearchForm&action_results=
[Accessed
23/01
2012].
NEA. 2004. The Energy Efficiency Commitment from April 2005 - Response by National
Energy Action [Online]. Available: http://www.nea.org.uk/assets/ConsultationPapers/2004/energy_efficiency_commitment_2005-2008.pdf [Accessed 23/01
2012].
NEA. 2006. Consultation Document Response. The Energy Efficiency Commitment April
2008 to March 2011: Initial Consultation [Online]. NEA. Available:
http://www.nea.org.uk/assets/ConsultationPapers/2006/The_Energy_Efficiency_Commitment_April_2008_to_March_2011.p
df [Accessed 23/01 2012].
NEA. 2007. Consultation Response Document. CERT - Carbon Emissions Reduction
Target 2008-2001 (EEC2 Replacement) [Online]. NEA. Available:
http://www.nea.org.uk/assets/ConsultationPapers/2007/Carbon_Emissions_Reduction_Target_Aug_2007_r_.pdf [Accessed
23/01 2012].
NEA. 2010. Extending the Carbon Emissions Reduction Target [Online]. NEA. Available:
http://www.nea.org.uk/consultation-paper-responses/results?selectyear=2010&formController=consultation-paperresponses%2F&executeForm=SearchForm&action_results=
[Accessed
21/01
2012].
NERA 2012. The Costs of the Energy Company Obligation. London: NERA.
Ness, E. C. 2010. The politics of determining merit aid eligibility criteria: An analysis of
the policy process. The Journal of higher education, 81, 33.
Neuerer, D. 2011. Ramsauers Energiewende torpediert Schäubles Finanzplanung.
Handelsblatt, 09/04/2011.
Newcomb, T. M. 1984. Conservation Program Evaluations. Evaluation Review, 8, 425440.
NIA. 2006. Submission by the National Insulation Association to the 2006 UK Energy
Review
[Online].
NIA.
Available:
http://www.nationalinsulationassociation.org.uk/media-andinformation/uploads%5Cnews%5Cid20%5C2006_Energy_Review_response.pdf
[Accessed 11/01 2012].
NIA. 2007. Response to the: “Carbon Emissions Reduction Target April 2008 to March
2011: Consultation Proposals May 2007” [Online]. NIA. Available:
http://www.nationalinsulationassociation.org.uk/media-andinformation/uploads%5Cnews%5Cid18%5CCERTResponse.pdf [Accessed 11/01
2012].
NIA. 2009. Amendments to the Carbon Emissions Reduction Target. Consultation
response from the National Insulation Association [Online]. NIA. Available:
http://www.nationalinsulationassociation.org.uk/media-andinformation/index.php?page=policies [Accessed 11/01 2012].
Nohrstedt, D. 2005. External shocks and policy change: Three Mile Island and Swedish
nuclear energy policy. Journal of European Public Policy, 12, 1041-1059.
298
Bibliography
Nohrstedt, D. 2008. The Politics of Crisis Policymaking: Chernobyl and Swedish Nuclear
Energy Policy. Policy Studies Journal, 36, 257.
North, D. 1990. Institutions, institutional change and economic performance, Cambridge,
Cambridge University Press.
North, D. 1998. Five propositions about institutional change. In: Knight, J. & Sened, I.
(eds.) Explaining social institutions. Ann Arbor: University of Michigan Press.
North, D. 2005. Understanding the process of economic change, Princeton, Oxford,
Princeton University Press.
Nowlin, M. C. 2011. Theories of the Policy Process: State of the Research and Emerging
Trends. Wiley-Blackwell.
NYSERDA 1 . New York’s System Benefits Charge Programs Evaluation and Status
Report. New York: NYSERDA.
OFFER 1991. Energy efficiency: consultation paper, Birmingham, OFFER.
OFFER 1992. Energy efficiency: the way forward, Birmingham, OFFER.
OFFER 1993. Energy Efficiency. Standards of Performance. A discussion paper
(unpublished).
OFFER 1994. Energy Efficiency: Standards of Performance. Birmingham: OFFER.
OFFER 1998a. Energy Efficiency Standards of Performance Public Electricity Suppliers
1998-2000. Birmingham: OFFER.
OFFER 1998b. Energy Efficiency: New Standards of Performance. Consultation Paper.
Birmingham: OFFER.
OFGAS 1994. The efficient use of gas: the role of OFGAS. London: OFGAS.
OFGEM 1999. Energy Efficiency: A Consultation Paper. London: OFGEM.
OFGEM 2000a. Energy efficiency Standards of performance 2000-2002. Final Decisions.
London: OFGEM.
OFGEM 2000b. Energy efficiency: Standards of performance 2000-2002. Final Decisions.
London: OFGEM.
OFGEM 2000c. The Social Action Plan. London: OFGEM.
OFGEM 2001. Energy Efficiency Commitment Administration Procedures. London:
OFGEM.
OFGEM 2002. Energy Efficiency Commitment. 2002-2005. Technical guidance manual
Issue 1. London: OFGEM.
OFGEM 2003. A review of the first year of the Energy Efficiency Commitment - A report
for the Secretary of State for Environment, Food and Rural Affairs. London:
OFGEM.
OFGEM 2004a. Energy Efficiency Commitment 2005 – 2008 Administration procedures.
London: OFGEM.
OFGEM 2004b. A review of the Energy Efficiency Commitment to the end of the second
year - A report for the Secretary of State for Environment, Food and Rural Affairs.
London: OFGEM.
OFGEM 2005. A review of the Energy Efficiency Commitment 2002 – 2005. London:
OFGEM.
OFGEM 2006. A review of the first year of the Energy Efficiency Commitment 2005
2008. London: OFGEM.
OFGEM 2007. A review of the second year of EEC 2005-2008. London: OFGEM.
OFGEM. 2008a. Energy Supply Markets Probe - Call for Evidence [Online]. Available:
http://www.ofgem.gov.uk/Markets/RetMkts/ensuppro/Documents1/Energy%20Sup
ply%20Markets%20Probe%20-%20Call%20for%20Evidence.pdf [Accessed 07/07
2010].
299
Bibliography
OFGEM. 2008b. Market is sound - Ofgem assures Chancellor [Online]. Available:
http://www.ofgem.gov.uk/Media/PressRel/Documents1/Ofgem%202.pdf
[Accessed 07/07 2010].
OFGEM 2008c. A review of the Energy Efficiency Commitment 2005-2008. London:
OFGEM.
OFGEM 2008d. What is the Carbon Emissions Reduction Target? CERT Update 1.
London: OFGEM.
OFGEM 2009a. Carbon Emissions Reduction Target (CERT) 2008-2011 Supplier
Guidance - Version 2. London: OFGEM.
OFGEM 2009b. Policy developments. CERT Update, 4.
OFGEM 2009c. A review of the first year of the Carbon Emissions Reduction Target.
London: OFGEM.
OFGEM 2010a. Carbon savings achieved. CERT Update 8. London: OFGEM.
OFGEM 2010b. A review of the second year of the Carbon Emissions Reduction Target.
London: OFGEM.
OFGEM 2011a. Carbon Emissions Reduction Target (CERT) 2008-2012 Supplier
Guidance - Version 3. London: OFGEM.
OFGEM 2011b. A review of the third year of the Carbon Emissions Reduction Target
(CERT). London: OFGEM.
OFGEM 2012a. A review of the fourth year 2012 of the Carbon Emissions Reduction
Target (CERT) programme 2008-2012. London: OFGEM.
OFGEM 2012b. Updated household energy bills explained. London: OFGEM.
OFGEM & Energy Saving Trust 2003. A review of the Energy Efficiency Standards of
Performance 1994 - 2002. London: OFGEM.
Oliver, J. 2008. Energy giants told to help poor or face levy. The Times. 02/03/2008 ed.
Ottnad, A. & Hefele, P. 2006. Die Zukunft der Bauwirtschaft in Deutschland. Umfeld Probleme - Perspektiven, München, Olzog Verlag.
Owen, G. 1995. Energy policy, the government and the energy regulators: a case study of
the Energy Saving Trust. Norwich: CSERGE.
Owen, G. 1997. Who controls energy efficiency policy? A case study of the energy saving
trust. Energy Policy, 25, 959.
Owen, G. 2006. Sustainable development duties: New roles for UK economic regulators.
Utilities policy, 14, 208.
Painter, J. 2011. Poles apart: the international reporting of climate scepticism, Oxford,
Reuters Institute for the Study of Journalism.
Patashnik, E. M. 2008. Reforms at risk : what happens after major policy changes are
enacted, Princeton, Oxford, Princeton University Press.
Pearce, D. 2006. The political economy of an energy tax: The United Kingdom's climate
change levy. Energy Economics, 28, 149-158.
Perl, A. & Dunn, J. A. 2007. Reframing Automobile Fuel Economy Policy in North
America: The Politics of Punctuating a Policy Equilibrium. Transport Reviews, 27,
1-35.
Peters, J. S. & McRae, M. 2008. Free-ridership measurement is out of sync with program
logic…or, we’ve got the structure built, but what’s its foundation? ACEEE Summer
Study on Energy Efficiency in Buildings. ACEEE.
Pettigrew, A. M. 1990. Longitudinal field research on change: theory and practice.
Organization Science, 1, 267.
Pickvance, C. 2009. The construction of UK sustainable housing policy and the role of
pressure groups. Local Environment, 14, 329-345.
300
Bibliography
Pierson, P. 1993a. When Effect Becomes Cause: Policy Feedback and Political Change.
World Politics, 45, 595-628.
Pierson, P. 1993b. When Effect Becomes Cause: Policy Feedback and Political Change.
World Politics, 45, 595-628.
Pierson, P. 1994. Dismantling the welfare state? Reagan, Thatcher, and the politics of
retrenchment, Cambridge, New York, Cambridge University Press.
Pierson, P. 2000a. Increasing returns, path dependence, and the study of politics. American
Political Science Review, 94, 251-267.
Pierson, P. 2000b. Not just what, but when: Timing and sequence in political processes.
Studies in American Political Development, 14, 72.
Pierson, P. 2001. The new politics of the welfare state. Oxford: Oxford University Press.
Pierson, P. 2004a. Politics in time : history, institutions, and social analysis, Princeton,
N.J. ; Oxford, Princeton University Press.
Pierson, P. 2004b. Politics in time: history, institutions, and social analysis, Princeton;
Oxford, Princeton University Press.
Pierson, P. 2006. Public Policies as Institutions. In: Shapiro, I., Skowronek, S. & Galvin,
D. (eds.) Rethinking political institutions : the art of the state. New York: New
York University Press.
Pierson, P. 2009. The study of policy development. Journal of policy history, 17, 34.
Pierson, P. & Skocpol, T. 2000. Historical Institutionalism in Contemporary Political
Science. In: Katznelson, I. & Milner, H. V. (eds.) Political Science: State of the
Discipline. New York: W. W. Norton.
Pierson, P. & Skocpol, T. 2002. Historical institutionalism in contemporary political
science. In: Katznelson, I. & Milner, H. V. (eds.) Political science: state of the
discipline. New York: W.W. Norton.
PIU 2002. The Energy Review. London: Performance and Innovation Unit, Cabinet
Office.
Powells, G. D. 2009. Complexity, entanglement, and overflow in the new carbon
economy: the case of the UK's Energy Efficiency Commitment. Environment and
Planning A, 41, 2342-2356.
Pralle, S. B. 2003. Venue shopping, political strategy, and policy change: the
internationalization of Canadian forest advocacy. Journal of Public Policy, 23, 233.
Preston, I., White, V. & Guertler, P. 2010. Distributional impacts of UK Climate Change
Policies. Final report to eaga Charitable Trust. Centre for Sustainable Energy and
Association for the Conservation of Energy.
Pump, B. 2011. Beyond Metaphors: New Research on Agendas in the Policy Process.
Policy Studies Journal, 39, 1-12.
Purple Market Research 2007. Trade Sector Profile. Domestic Insulation 2007. Prepared
for EST and EEPH. London: EST and EEPH.
Purple Market Research 2009. Solid Wall Insulation Supply Chain Review. Prepared for
EST and EEPH. London: EST and EEPH.
Pyett, P. M. 2003. Validation of Qualitative Research in the" Real World". Qualitative
health research, 13, 1170.
Ramsauer, P. 2010. Letter to the coaltion MPs of the CDU/CSU and FDP. Berlin.
Ramsauer, P. 2011. Pressekonferenz zum Energiekonzept der Bundesregierung mit
Bundeskanzlerin Merkel, BM Rösler, BM Röttgen und BM Ramsauer [Online].
Available:
http://www.bundesregierung.de/Content/DE/Mitschrift/Pressekonferenzen/2011/05
/2011-05-30-pk-bk-bm-energiekonzept.html [Accessed 06/03 2012].
301
Bibliography
Rayner, J. 2009. Understanding Policy Change as a Historical Problem. Journal of
Comparative Policy Analysis, 11, 83.
Reimer, N. 2003. Bundesbau- und Bundesumweltminister unterzeichnen neuen
Fördervertrag zur klimarelevanten Gebäudesanierung. Einsparpotenzial von 40
Millionen Tonnen CO2. Studie des WWF: Stromwirtschaft könnte bis 2020 ihren
Ausstoß halbieren. taz, p.8.
Renzsch, W. 2010. Die zweite Stufe der Bundesstaatsreform. In: Egle, C. & Zohlnhöfer,
R. (eds.) Die zweite Große Koalition: eine Bilanz der Regierung Merkel 20052009. 1. Aufl. ed. Wiesbaden: VS Verlag für Sozialwissenschaften.
Rheinische Post. 2010. Minister streiten um Hilfen für Hausbesitzer. Available:
http://nachrichten.rp-online.de/wirtschaft/minister-streiten-um-hilfen-fuerhausbesitzer-1.98294 [Accessed 20/03/2012].
Ringer, F. K. 2004. Max Weber: an intellectual biography, Chicago; London, University
of Chicago Press.
Rose, R. 1976. Models of change. In: Rose, R. (ed.) The Dynamics of public policy: a
comparative analysis. London: Sage.
Rosenow, J. Different Paths of Change: Home Energy Efficiency Policy in Britain and
Germany. ECEEE Summer Study 2011, 2011 Belambra Presqu'île de Giens,
France. 261-272.
Rosenow, J. 2012. Energy Savings Obligations in the UK – A History of Change. Energy
Policy, 49, 373–382.
Rosenow, J. in press. The Politics of the German CO2 Building Rehabilitation Programme.
Energy Efficiency.
Rosenow, J. & Eyre, N. 2012. Green Deal and the Energy Company Obligation - will it
work? British Institute of Energy Economics 9th Academic Conference. European
Energy in a Challenging World: The impact of emerging markets. Oxford.
Rosenow, J., Platt, R. & Flanagan, B. 2012. Fuel poverty and energy savings obligations:
Natural bedfellows or enemies? Fuel poverty in policy and practice - a
postgraduate symposium. University of Sheffield.
Royal Commission on Environmental Pollution 2000. Energy - The Changing
Environment, London, TSO.
RWI forsa
8. Erhebung des Energieverbrauchs der privaten Haushalte f r das Jahr
2005.
Sabatier, P. A. 1988. An advocacy coalition framework of policy change and the role of
policy-oriented learning therein. Policy Sciences, 21, 129.
Sabatier, P. A. 1993. Policy change over a decade or more. In: Sabatier, P. A. & JenkinsSmith, H. C. (eds.) Policy change and learning : an advocacy coalition approach.
Boulder, Colo.: Westview Press.
Sabatier, P. A. 1998. The advocacy coalition framework: revisions and relevance for
Europe. Journal of European Public Policy, 5, 98 - 130.
Sabatier, P. A. 2007a. Fostering the development of policy theory. In: Sabatier, P. A. (ed.)
Theories of the policy process. 2nd ed. Boulder, Colo.: Westview Press.
Sabatier, P. A. 2007b. Theories of the policy process, Boulder, Colo., Westview Press.
Sabatier, P. A. & Jenkins-Smith, H. C. 1993. Policy change and learning : an advocacy
coalition approach, Boulder, Colo., Westview Press.
Sabatier, P. A. & Jenkins-Smith, H. C. 1999. The Advocacy Coalition Framework: An
Assessment. In: Sabatier, P. A. (ed.) Theories of the policy process. Boulder, Colo.:
Westview Press.
302
Bibliography
Sabatier, P. A. & Weible, C. M. 2007. The Advocacy Coalition Framework: Innovations
and Clarifications. In: Sabatier, P. A. (ed.) Theories of the policy process. 2nd ed.
Boulder: Westview Press.
Saint-Germain, M. A. 1996. Three strikes and you're in: A streams and windows model of
incremental policy change. Journal of Criminal Justice, 24, 57.
Sanders, C. & Phillipson, M. 2006. Review of Differences between Measured and
Theoretical Energy Savings for Insulation Measures. Glasgow: Glasgow
Caledonian University.
Sant, R. W. 1979. The least-cost energy strategy: minimizing consumer costs through
competition, Arlington, Va, Mellon Institute.
Schäfer, J. W. 2010. Experten warnen vor drastischen Mieterhöhungen. Bild [Online].
Available:
http://www.bild.de/politik/wirtschaft/wirtschaftspolitik/kosten-jedesjahr-erhoeht-13977996.bild.html [Accessed 20/03/2012].
Schäfer, M. S., Ivanova, A. & Schmidt, A. 2011. Media Attention for Climate Change in
Germany [Online]. Hamburg: KlimaCampus, University of Hamburg. Available:
http://www.klimacampus.de/macc.html.
Schattschneider, E. E. 1935. Politics, pressures and the tariff : a study of free private
enterprise in pressure politics, as shown in the 1929-1930 revision of the tariff,
New York, Prentice-Hall, inc.
Schlager, E. 1995. Policy making and collective action: defining coalitions within the
advocacy coalition framework. Policy Sciences, 28, 243.
Schlager, E. 1999. A Comparison of Frameworks, Theories, and Models of Policy
Processes. In: Sabatier, P. A. (ed.) Theories of the policy process. Boulder, Colo.:
Westview Press.
Schlomann, B. & Eichhammer, W. 2012. Energy Efficiency Policies and Measures in
Germany in 2012. Karlsruhe: Fraunhofer Institute for Systems and Innovation
Research ISI.
Schrad, M. L. 2010. The political power of bad ideas: networks, institutions, and the
global prohibition wave, New York; Oxford, Oxford University Press.
Schröder, F., Altendorf, L., Greller, M. & Boegelein, T. 2011. Universelle
Energiekennzahlen für Deutschland: Teil 4: Spezifischer Heizenergieverbrauch
kleiner
Wohnhäuser
und
Verbrauchshochrechnung
für
den
Gesamtwohnungsbestand. Bauphysik, 33, 243-253.
Schroeder, M., Ekins, P., Power, A., Zulauf, M. & Lowe, R. J. 2011. The KFW experience
in the reduction of energy use in and CO2 emissions from buildings: operation,
impacts and lessons for the UK. London: UCL Energy Institute, University College
London and LSE Housing and Communities, London School of Economics.
Sch le, R.
Hanke, T.
6. Evaluation des Online-Modernisierungsratgebers von
co2online. Wuppertal: Wuppertal Institute.
ScottishPower 2007. Carbon Emissions Reduction Target April 2008 - March 2011.
Consultation response by ScottishPower. Glasgow: ScottishPower.
Sharratt, D., Brigham, B. H. & Brigham, M. 2007. The Utility of Social Obligations in the
UK Energy Industry. Journal of Management Studies, 44, 1503-1522.
Simon, H. A. 1957. Models of Man: Social and Rational. New York: Wiley.
Simon, H. A. 1959. Theories of decision-making in economics and behavioral science.
American Economic Review, 49, 253-283.
Simon, H. A. 1983. Reason in human affairs, Oxford, Basil Blackwell.
Skea, J. 2012. Research and evidence needs for decarbonisation in the built environment: a
UK case study. Building Research & Information, 40, 432-445.
303
Bibliography
Skocpol, T. 1979. States and social revolutions: a comparative analysis of France, Russia
and China, Cambridge, Cambridge University Press.
Skocpol, T. 1992. Protecting soldiers and mothers: the political origins of social policy in
the United States, Cambridge, Mass.; London, Belknap Press of Harvard
University Press.
Smith, A. 2000. Policy networks and advocacy coalitions: explaining policy change and
stability in UK industrial pollution policy? Environment and planning. C,
Government & policy, 18, 95.
Sommer, J. & Rosenthal, P. 2012. Active labour market policy and direct job creation in
Germany: Achievements, challenges, and pitfalls. Local Economy, 27, 494-501.
Sorrell, S. 2007. The rebound effect: an assessment of the evidence for economy-wide
energy savings from improved energy efficiency. London: UK Energy Research
Centre.
Spiegel. 2010. Unionspolitiker zerpflücken Ramsauers Kürzungspläne [Online]. Available:
http://www.spiegel.de/wirtschaft/soziales/0,1518,718126,00.html [Accessed 20/03
2012].
SSE 2007. SSE Response to consultation on Carbon Emissions Reduction Target April
2008 - March 2011. Reading: SSE.
Statistisches Bundesamt 2008. Schulden der öffentlichen Haushalte. Fachserie 14 Reihe 5.
2007. Wiesbaden: Statistisches Bundesamt.
Statistisches Bundesamt 2010. Schulden des öffentlichen Gesamthaushaltes 2009 im
Zeichen der Finanzmarkt und Wirtschaftskrise. Wirtschaft und Statistik, 8, 754764.
Statistisches Bundesamt 2011. Schulden der öffentlichen Haushalte. Fachserie 14 Reihe 5.
2010. Wiesbaden: Statistisches Bundesamt.
Statistisches Bundesamt. 2012a. Bruttoinlandsprodukt preisbereinigt, Kettenindex (2005 =
100)
[Online].
Available:
http://www.destatis.de/jetspeed/portal/cms/Sites/destatis/Internet/DE/Content/Statis
tiken/Zeitreihen/WirtschaftAktuell/VolkswirtschaftlicheGesamtrechnungen/Conten
t100/kvgr111x12.psml [Accessed 14/02 2012].
Statistisches Bundesamt. 2012b. Energieverbrauch der privaten Haushalte für Wohnen
stark
rückläufig
[Online].
Available:
https://www.destatis.de/DE/ZahlenFakten/GesamtwirtschaftUmwelt/Umwelt/Umw
eltoekonomischeGesamtrechnungen/EnergieRohstoffeEmissionen/Aktuell.html
[Accessed 04/12 2012].
Statistisches Bundesamt. 2012c. Tätige Personen, Entgelte, Umsatz im Bauhauptgewerbe
(alle Betriebe) 1995-2011 [Online]. Statistisches Bundesamt. Available:
https://wwwgenesis.destatis.de/genesis/online?sequenz=tabelleAufbau&selectionname=441110001&sprache=de [Accessed 15/02 2012].
Steinmo, S. 1989. Political Institutions and Tax Policy in the United States, Sweden, and
Britain. World Politics, 41, 500-535.
Steinmo, S., Thelen, K. A. & Longstreth, F. 1992. Structuring politics: historical
institutionalism in comparative analysis, Cambridge, Cambridge University Press.
Stemler, S. 2001. An overview of content analysis. Practical Assessment, Research &
Evaluation, 7.
Stern, N. 2007. The Economics of Climate Change: The Stern Review, Cambridge,
Cambridge University Press.
304
Bibliography
Stratton, A. 2008. 70 MPs sign petition for windfall tax on energy and utility firms to help
fuel poor. Guardian. 25/08/2008 ed.
Streeck, W. 2010. Noch so ein Sieg, und wir sind verloren. Der Nationalstaat nach der
Finanzkrise. Leviathan, 38, 159-173.
Streeck, W. & Thelen, K. A. 2005. Beyond continuity : institutional change in advanced
political economies, Oxford, Oxford University Press.
Süddeutsche Zeitung. 2008. SPD legt eigenes Konjunkturprogramm vor [Online].
Available: http://www.sueddeutsche.de/politik/wirtschaftskrise-spd-legt-eigeneskonjunkturprogramm-vor-1.531990 [Accessed 01/02 2012].
Sunikka-Blank, M. & Galvin, R. 2012. Introducing the prebound effect: relationships
between energy performance ratings and actual heating energy consumption in
German dwellings, and their policy implications. Building Research and
Information, 40, 260-273.
Tansey, O. 2007. Process Tracing and Elite Interviewing: A Case for Non-probability
Sampling. PS: Political Science & Politics, 40, 765-772.
Thatcher, M. 1988. Speech to the Royal Society [Online]. Available:
http://www.margaretthatcher.org/document/107346 [Accessed 12/12 2011].
Thelen, K. A. & Steinmo, S. 1992. Historical Institutionalism in Comparative Politics. In:
Steinmo, S., Thelen, K. A. & Longstreth, F. (eds.) Structuring politics : historical
institutionalism in comparative analysis. Cambridge: Cambridge University Press.
Thomsen, K. E., Schultz, J. M. & Poel, B. 2005. Measured performance of 12
demonstration projects—IEA Task 13 “advanced solar low energy building.
Energy and Buildings, 37.
Tighelaar, C. & Menkveld, M. 2011. Obligations in the existing housing stock: Who pays
the bill? ECEEE 2011 Summer Study: Energy efficiency first: The foundation of a
low-carbon society. Belambra Presqu'île de Giens, France.
Toke, D. 2000. Policy network creation: the case of energy efficiency. Public
administration, 78, 835.
Toke, D. 2005. Are green electricity certificates the way forward for renewable energy?
An evaluation of the United Kingdom's Renewables Obligation in the context of
international comparisons. Environment and planning. C, Government & policy,
23, 361.
Toke, D. & Lauber, V. 2007. Anglo-Saxon and German approaches to neoliberalism and
environmental policy: The case of financing renewable energy. Geoforum, 38, 677687.
Train, K. E. 1994. Estimation of net savings from energy-conservation programs. Energy,
19, 423.
True, J. L. 2000. Avalanches and Incrementalism. The American Review of Public
Administration, 30, 3-18.
True, J. L., Jones, B. D. & Baumgartner, F. R. 2007a. Punctuated-Equilibrium Theory.
Explaining Stability and Change in Public Policymaking. In: Sabatier, P. A. (ed.)
Theories of the policy process. Boulder: Westview Press.
True, J. L., Jones, B. D. & Baumgartner, F. R. 2007b. Punctuated-Equilibrium Theory:
Explaining Stability and Change in Public Policymaking. In: Sabatier, P. A. (ed.)
Theories of the policy process. 2nd ed. Boulder, Colo.: Westview Press.
UBA 2012. Entwicklung der spezifischen Kohlendioxid-Emissionen des deutschen
Strommix 199 - 1 und erste Sch tzungen 11. Dessau-Roßlau: UBA.
Unruh, G. C. 2000. Understanding carbon lock-in. Energy Policy, 28, 817.
305
Bibliography
van Asche, T. 2012. When do new issues appear? Punctuations in the Belgian executive
agenda. Acta politica, 47, 128-150.
Varone, F. & Aebischer, B. 2001. Energy efficiency: the challenges of policy design.
Energy Policy, 29, 615-629.
Verbong, G. & Geels, F. 2007. The ongoing energy transition: Lessons from a sociotechnical, multi-level analysis of the Dutch electricity system (1960–2004). Energy
Policy, 35, 1025-1037.
Vickrey, W. S. 1969. Congestion Theory and Transport Investment. The American
Economic Review, 59, 251-260.
Vine, E., Hall, N., Keating, K., Kushler, M. & Prahl, R. 2012. Emerging issues in the
evaluation of energy-efficiency programs: the US experience. Energy efficiency, 5,
5-17.
Vine, E. L. 1994. The human dimension of program evaluation. Energy, 19, 165-178.
Vine, E. L. & Sathaye, J. A. 2000. The monitoring, evaluation, reporting, verification, and
certification of energy-efficiency projects. Mitigation and Adaptation Strategies for
Global Change, 5, 189.
Walgrave, S. & Varone, F. 2008. Punctuated Equilibrium and Agenda-Setting: Bringing
Parties Back in: Policy Change after the Dutroux Crisis in Belgium. Governance,
21, 365-395.
Walgrave, S. & Vliegenthart, R. 2010. Why are policy agendas punctuated? Friction and
cascading in parliament and mass media in Belgium. Journal of European Public
Policy, 17, 1147-1170.
Weaver, K. 2010. Paths and Forks or Chutes and Ladders?: Negative Feedbacks and
Policy Regime Change. Journal of Public Policy, 30, 137-162.
Webb, T. 2008. Windfall tax being blown off course. Guardian. 31/08/2008 ed.
Weber, R. P. 1990. Basic content analysis, Newbury Park ; London, Sage.
Webster, P. & Elliott, F. 2008. Gordon Brown reluctant on energy windfall tax. The Times.
25/08/2008 ed.
Weible, C. M., Sabatier, P. A. & McQueen, K. 2009. Themes and Variations: Taking
Stock of the Advocacy Coalition Framework. Policy Studies Journal, 37, 121.
Weir, M. 1992. Ideas and the Politics of Bounded Innovation. In: Steinmo, S., Thelen, K.
A. & Longstreth, F. (eds.) Structuring politics: historical institutionalism in
comparative analysis. Cambridge: Cambridge University Press.
Weiss, A. & Woodhouse, E. 1992. Reframing incrementalism: A constructive response to
the critics. Policy Sciences, 25, 255-273.
Wengraf, T. 2001. Qualitative research interviewing : semi-structured, biographical and
narrative methods, Thousand Oaks, Calif., London, SAGE.
Whittemore, R. 2001. Validity in qualitative research. Qualitative health research, 11,
522.
Wighton, A. 2008. Windfall tax is no easy option. The Times. 05/08/2008 ed.
Williams, R. A. 2009. Exogenous Shocks in Subsystem Adjustment and Policy Change:
The Credit Crunch and Canadian Banking Regulation. Journal of Public Policy,
29, 29.
Wintour, P. 2008. Green energy help for poor: Poorest targeted with energy-saving
schemes. Guardian. 04/08/2008 ed.
Wirl, F. 2000. Lessons from utility conservation programs. The Energy Journal, 21, 87.
Wood, G. & Dow, S. 2011. What lessons have been learned in reforming the Renewables
Obligation? An analysis of internal and external failures in UK renewable energy
policy. Energy Policy, 39, 2228-2244.
306
Bibliography
Wood, R. S. 2006. Tobacco's Tipping Point: The Master Settlement Agreement as a
Focusing Event. Policy Studies Journal, 34, 419-436.
Woodman, B. & Mitchell, C. 2011. Learning from experience? The development of the
Renewables Obligation in England and Wales 2002–2010. Energy Policy, 39,
3914-3921.
Woolf, T. 1993. Energy efficiency in Britain: creating profitable alternatives. Utilities
policy, 3, 233.
Workman, S., Jones, B. D. & Jochim, A. E. 2009. Information Processing and Policy
Dynamics. Policy Studies Journal, 37, 75-92.
Wormald, B. H. G. 1993. Francis Bacon : history, politics and science, 1561-1626,
Cambridge, Cambridge University Press.
Wurzel, R. 2008. The Politics of Emissions Trading in Britain and Germany. Berlin:
Anglo-German Foundation.
Zahariadis, N. 1992. To Sell or Not to Sell? Telecommunications Policy in Britain and
France. Journal of Public Policy, 12, 355-376.
Zahariadis, N. 1995. Markets, states, and public policy: privatization in Britain and
France, Ann Arbor, University of Michigan Press.
Zahariadis, N. 1996. Selling British Rail: An Idea Whose Time Has Come? Comparative
Political Studies, 29, 400.
Zahariadis, N. 1998. Comparing three lenses of policy choice. Policy Studies Journal, 26,
434.
Zahariadis, N. 1999. Ambiguity, Time, and Multiple Streams. In: Sabatier, P. A. (ed.)
Theories of the policy process. Boulder, Colo.: Westview Press.
Zylka, R. 2008. "Wir handeln und setzen Wachstumsimpulse" [Online]. Berlin: Berliner
Zeitung. Available: http://www.berliner-zeitung.de/archiv/spd-fraktionschef-peterstruck-ueber-die-folgen-der-finanzkrise-und-die-reform-der-erbschaftsteuer--wirhandeln-und-setzen-wachstumsimpulse-,10810590,10596522.html
[Accessed
15/02 2012].
307
Appendix
Appendix
List of Interviewees
UK
INDEX NUMBER
POSITION
DATE
I_UK1
Senior Energy Researcher
05/05/2011
I_UK2
Civil Servant at DECC
25/05/2011
I_UK3
Former Senior DETR Official
31/05/2011
I_UK4
Former Director of UK Energy Agency
09/06/2011
I_UK5
Official at OFGEM
20/06/2011
I_UK6
Civil Servant at DECC
21/06/2011
I_UK7
Representative of the Energy Efficiency Industry
23/06/2011
I_UK8
Independent Policy Adviser
28/06/2011
I_UK9
Former Energy Efficiency Manager at an Energy Supplier
05/07/2011
I_UK10
Former Director General at Energy Efficiency Office
05/07/2011
I_UK11
Representative of the Energy Efficiency Industry
14/09/2011
I_UK12
Senior Manager at Energy Supplier I
26/09/2011
I_UK13
Senior Manager at Energy Supplier II
06/10/2011
I_UK14
Fuel Poverty Expert, Consumer Focus
27/11/2012
308
Appendix
Germany
INDEX NUMBER
POSITION
DATE
I_G1
Energy research institute, Researcher
05/08/2011
I_G2
NABU, Energy Efficiency Expert
22/08/2011
I_G3
BMWi, Civil Servant
23/08/2011
I_G4
KfW Banking Group, Member of Staff
28/09/2011
I_G5
DENA, Member of Staff
27/09/2011
I_G6
BMVBS, Civil Servant
24/10/2011
I_G7
Researcher for Member of Parliament, Green Party
14/11/2011
I_G8
Member of Parliament, SPD
15/12/2011
I_G9
Member of Parliament, FDP
14/12/2011
I_G10
Member of Parliament, CDU
15/12/2011
I_G11
Researcher for Member of Parliament, The Left
15/12/2011
I_G12
Member of Parliament, Green Party
19/12/2011
309